Add files using upload-large-folder tool
Browse files- README.md +245 -0
- meta/train/chunk_00000.parquet +3 -0
- meta/validation/chunk_00000.parquet +3 -0
- parquet_info.json +10 -0
- preview/train/chunk_00000.parquet +3 -0
- preview/validation/chunk_00000.parquet +3 -0
- reports/build/counters.json +99 -0
- reports/build/dropped.jsonl +0 -0
- reports/build/manifest.json +49 -0
- reports/build/quarantine.jsonl +61 -0
- reports/export_manifest.json +261 -0
- reports/known_issues/readback_counts_beyond_target.tsv +8 -0
- reports/known_issues/readback_excludes_same_category.tsv +29 -0
- reports/known_issues/summary.json +20 -0
- reports/s10_dropped.jsonl +0 -0
- reports/s10_label_conflicts.jsonl +0 -0
- reports/s13_dropped.jsonl +302 -0
- reports/s14_sample.tsv +201 -0
- train/chunk_00000.parquet +3 -0
- validation/chunk_00000.parquet +3 -0
README.md
ADDED
|
@@ -0,0 +1,245 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
license_name: upstream-undeclared
|
| 4 |
+
license_link: https://huggingface.co/datasets/leo66666/messytable
|
| 5 |
+
pretty_name: "Spatial MMCoT v1 - messytable"
|
| 6 |
+
language:
|
| 7 |
+
- en
|
| 8 |
+
task_categories:
|
| 9 |
+
- visual-question-answering
|
| 10 |
+
- image-to-image
|
| 11 |
+
tags:
|
| 12 |
+
- spatial-reasoning
|
| 13 |
+
- interleaved-reasoning
|
| 14 |
+
- multimodal-chain-of-thought
|
| 15 |
+
size_categories:
|
| 16 |
+
- 1K<n<10K
|
| 17 |
+
configs:
|
| 18 |
+
- config_name: preview
|
| 19 |
+
default: true
|
| 20 |
+
data_files:
|
| 21 |
+
- split: train
|
| 22 |
+
path: preview/train/*.parquet
|
| 23 |
+
- split: validation
|
| 24 |
+
path: preview/validation/*.parquet
|
| 25 |
+
- config_name: train
|
| 26 |
+
data_files:
|
| 27 |
+
- split: train
|
| 28 |
+
path: train/*.parquet
|
| 29 |
+
- split: validation
|
| 30 |
+
path: validation/*.parquet
|
| 31 |
+
- config_name: meta
|
| 32 |
+
data_files:
|
| 33 |
+
- split: train
|
| 34 |
+
path: meta/train/*.parquet
|
| 35 |
+
- split: validation
|
| 36 |
+
path: meta/validation/*.parquet
|
| 37 |
+
---
|
| 38 |
+
|
| 39 |
+
# Spatial MMCoT v1 · `messytable`
|
| 40 |
+
|
| 41 |
+
Multi-camera tabletop counting on real photographs from MessyTable (Z. Cai, J. Zhang, D. Ren, C. Yu, H. Zhao, S. Yi, C. K. Yeo and C. C. Loy, "MessyTable: Instance Association in Multiple Camera Views", ECCV 2020; https://github.com/caizhongang/MessyTable). The questions, counts and reasoning text come from `leo66666/messytable`, which pairs those photographs with a counting chain of thought. Several angled views of one cluttered table are the input; the plan notes that angled views double-count under occlusion; the target is the real photograph from MessyTable's overhead camera (cam1), which never appears among the inputs (when the read-back says 'the generated top-down view' it means the view the model is trained to produce); the read-back counts off it. The overhead frame is a crop, though, and on a few rows the read-back adds objects it says are outside or hidden in that frame, taking them from the angled views. The answer is upstream's `gt_answer`, and on questions about a broad category ('canned beverage/food', 'opaque cup', ...) it is often lower than the number of such objects on the table (see Known issues). Only the upstream train split is used, and both splits here, validation included, are carved from it by capture scene. The upstream test split is the pool from which the IPT paper's MessyTable counting benchmark (MVC_MessyTable_ImaginativePerceptionToken) draws its items. This release shares no capture scene with that split, but it has the same camera rig, object inventory and question templates: a model trained on this source is in-domain for that benchmark and should not report it as out-of-distribution. If you use these images, cite MessyTable.
|
| 42 |
+
|
| 43 |
+
**Supervision kind** (`supervision_kind` in `meta`): `full_interleaved` on every row: the upstream trace itself interleaves text and target images (drawn or rendered states on most sources; the source note above says which), and the read-back comes after the target image it reads. The text is upstream's and was not checked against the images.
|
| 44 |
+
|
| 45 |
+
Upstream: [`leo66666/messytable`](https://huggingface.co/datasets/leo66666/messytable). Licence: **undeclared**. The upstream repository declares no licence; this converted copy is shared for research use only, whatever terms the upstream authors set apply to it as well, and it will be taken down at their request. The photographs are MessyTable's: its code repository is MIT-licensed, and neither that repository nor the project page states separate terms for the image data.
|
| 46 |
+
|
| 47 |
+
## Known issues
|
| 48 |
+
|
| 49 |
+
Rows with a measured per-row problem are listed in `reports/known_issues/`, one TSV per issue (a `# <description>` line, then `row_uid<TAB>split<TAB>detail` lines), so they can be filtered out. They are still in this release: no row was removed for these issues.
|
| 50 |
+
|
| 51 |
+
| issue | rows | train | validation | what | how it was found | file |
|
| 52 |
+
|---|---:|---:|---:|---|---|---|
|
| 53 |
+
| `readback_excludes_same_category` | 28 | 27 | 1 | The question names a broad category, and the read-back reaches the (lower) label by setting aside objects of that category as a different type/brand/style ('The two orange cups are also opaque but are a different style'), by naming only the 'boxed snacks of interest', or by counting groups as one ('treating each group as one'); upstream's gt_answer appears to count one product, not the whole category. | the read-back says 'of interest', 'treating each group' or 'this scene's labeling', or it has an exclusion phrase ('different type|kind|brand|style|product|item|drink|packaging|container', 'not counted / not the queried ... category|type', 'specified') whose subject (from the sentence start, or from an unclosed '(' that opens its own clause, to the phrase) names an object of the queried category (its non-generic words, or their container noun: bottle, can, box/carton, bag/pouch/packet, cup/mug/tumbler); measured 2026-09-25; known_issues.py sha1 655e5307; export 4aaee1f rows be1d3dc7/6070fad9 | `reports/known_issues/readback_excludes_same_category.tsv` |
|
| 54 |
+
| `readback_counts_beyond_target` | 7 | 7 | 0 | The read-back does not count off the overhead target alone: it adds objects it says the angled input views show but the overhead crop does not (a second pineapple outside the crop, a stack seen from the side; on at least one row the added object is in no image). | a read-back sentence mentions the angled/input/original/side views and adds something (additional, extra, another, one more, a second, bringing the total, not visible in the top-down view, outside the crop, a stack of two) with no negation (no/not/without/any/rather than) in the 30 characters before it, not in a clause whose verb is 'confirm(s)' unless it confirms 'an additional/another ...', and not 'appear(s) as additional'; measured 2026-09-25; known_issues.py sha1 655e5307; export 4aaee1f rows be1d3dc7/6070fad9 | `reports/known_issues/readback_counts_beyond_target.tsv` |
|
| 55 |
+
|
| 56 |
+
To leave the listed rows out (the snippet in the loader section downloads `reports/known_issues/` with the data):
|
| 57 |
+
|
| 58 |
+
```python
|
| 59 |
+
import glob, os
|
| 60 |
+
root = "<root>/messytable"
|
| 61 |
+
drop = {line.split("\t")[0] for f in glob.glob(os.path.join(root, "reports/known_issues/*.tsv"))
|
| 62 |
+
for line in open(f) if line.strip() and not line.startswith(("#", "row_uid\t"))}
|
| 63 |
+
# keep a row when its row_uid (a column of train/, meta/ and preview/) is not in drop
|
| 64 |
+
```
|
| 65 |
+
|
| 66 |
+
### Measured caveats
|
| 67 |
+
|
| 68 |
+
Measured on this release by the pre-publication review (2026-09-25): problems that cannot be listed row by row (a shortcut in the options, a label convention, an upstream labelling scheme) and what the review found around the lists above. Where a caveat counts listed rows ("listed as `...`"), the count is the table's, read from `reports/known_issues/summary.json`. Its other numbers are the review's own measurements, which no file carries: they hold for exactly these rows and are not re-measured automatically. Items marked **Training-signal defect** are problems in what the rows teach, not only in how they are described; no row was removed for them.
|
| 69 |
+
|
| 70 |
+
- **Training-signal defect.** The answer is upstream's `gt_answer`, and on questions about a broad category it is often lower than the number of such objects on the table. The questions name a category ('canned beverage/food', 'bottled beverage', 'boxed snack', 'instant cup meal', 'opaque cup', ...); where the table holds several products of that category, the label appears to count only one of them, and no label is above 8. For example `3679ee397a0860d4` asks 'Count canned beverage/food in this scene.' and the answer is 5, while its overhead photograph shows about 22 cans in five groups. Of 20 randomly drawn training rows about such categories, at least 4 plainly show more objects of the named category than the answer, and about 8 more probably do. The read-backs were written to reach the stored answer, so on these rows they leave out the other products, exclude them outright ('The two orange cups are also opaque but are a different style', `36b8d38bf3a3396e`) or regroup them ('Treating each group as one distinct canned-beverage/food instance', `3679ee397a0860d4`); the 28 rows (27 train, 1 validation; listed as `readback_excludes_same_category`) are the read-backs that say so in words. A model trained on these rows learns to under-count a named category. All 61 S4c quarantines point the same way (each read-back counted more than the label), so the rows kept lean toward read-backs that agree with the narrower count. Read the answer as the count of one product, not of the whole category. The upstream test split, from which the IPT MessyTable benchmark draws, is built the same way.
|
| 71 |
+
- The overhead frame is a crop (`union_center`) and does not always show every counted object: on 7 rows (7 train, 0 validation; listed as `readback_counts_beyond_target`), 0.5% of the rows, the read-back adds objects it says are outside or hidden in that frame, taking them from the angled views (e.g. `37eff3f7b91a3eb2`, a second pineapple outside the crop); at least one such object is in no image (`9ba7d5ebb7e52811`), and a few read-backs' own arithmetic does not match their stated total. 122 of the 1,517 read-backs appeal to the angled views.
|
| 72 |
+
|
| 73 |
+
## Size
|
| 74 |
+
|
| 75 |
+
| split | rows | target image slots | distinct target images |
|
| 76 |
+
|---|---:|---:|---:|
|
| 77 |
+
| train | 1,490 | 1,490 | 1,490 |
|
| 78 |
+
| validation | 27 | 27 | 27 |
|
| 79 |
+
|
| 80 |
+
| task | train | validation |
|
| 81 |
+
|---|---:|---:|
|
| 82 |
+
| multiview_counting | 1,490 | 27 |
|
| 83 |
+
|
| 84 |
+
Input images per row: 2 to 7. Target images per row (the images the model is trained to generate): 1.
|
| 85 |
+
Image corpus (`source_scene_corpus`): messytable 1,517.
|
| 86 |
+
|
| 87 |
+
## Row format
|
| 88 |
+
|
| 89 |
+
One row is: input image(s) and a question, then K rounds of *thought → target image* (the target is the source's own ground-truth image, which the model is trained to generate), then a final thought (normally a read-back of the last target; where a source's final thought is something else, or often leaves out the answer, the source note or Known issues says so) and the answer; here K is 1. In the `train` config:
|
| 90 |
+
|
| 91 |
+
```
|
| 92 |
+
image_list list<binary> inputs first, then the K target images in order
|
| 93 |
+
num_input_images int64 how many of image_list are inputs
|
| 94 |
+
instruction_list list<string> one element: system prompt + question
|
| 95 |
+
output_text_list list<string> K+1 elements:
|
| 96 |
+
[0] <think>plan 1</think><image_start>
|
| 97 |
+
[j] <image_end><think>plan j+1</think><image_start>
|
| 98 |
+
[K] <image_end><think>read-back</think><answer>answer</answer>
|
| 99 |
+
row_uid string join key to `meta` and `preview`
|
| 100 |
+
```
|
| 101 |
+
|
| 102 |
+
Every image is a JPEG, and no input image is larger than 512 px on its long edge (measured on this release, 2026-09-25); the size each target was stored at is `target_px` in `meta`.
|
| 103 |
+
|
| 104 |
+
`<answer>` holds exactly `meta.answer_value` (also the `answer` column of `preview`) on every row: score model output against that string.
|
| 105 |
+
|
| 106 |
+
The system prompt is ThinkMorph's `VLM_THINK_SYSTEM_PROMPT` from its `inferencer.py`, verbatim (`GEN_THINK_SYSTEM_PROMPT` there has the same text), including its leading and trailing newline. The markers are plain strings, not tokenizer special tokens; the prompt writes `</image_end>` and the data writes `<image_end>`, exactly as the ThinkMorph-7B checkpoint was trained.
|
| 107 |
+
|
| 108 |
+
`preview` shows the same rows with one column per slot: `input_image_i` for the inputs; for each of the K = `num_steps` rounds, the plan `thought_j` and its target `target_image_j`; and the read-back in `thought_1` on every row.
|
| 109 |
+
|
| 110 |
+
`meta` holds the per-row sidecar: task, `scene_id` and `geometry_uid` (the scene and geometry keys; the split key is named in the split paragraph below), `trajectory_id` (a camera-path or sample label, empty where the source has none), `num_steps`, `num_input_images`, `answer_type`, `answer_value`, `majority_class_rate`, `target_image_kind`, `target_px`, `est_tokens`, licence, `split` (`train` / `validation`, the Hub split names), `supervision_kind` (`full_interleaved` / `visual_aux` / `visual_only`) and `filter_flags`. `majority_class_rate` is the share of the task's most frequent `answer_value` among its training rows: it measures answer skew and is not a guessing baseline (where a task mixes question types or each row has its own options it can be far below chance); compare scores with the text-only baselines below.
|
| 111 |
+
|
| 112 |
+
Per-row `license` in `meta`: undeclared 1,517.
|
| 113 |
+
|
| 114 |
+
Flags on released rows (`filter_flags` in `meta` and `preview`, comma-separated):
|
| 115 |
+
|
| 116 |
+
| flag | rows | meaning |
|
| 117 |
+
|---|---:|---|
|
| 118 |
+
| `S5.replay_unsupported` | 1,517 | no solver re-derives this task's answer from the trace, so S5 did not replay it |
|
| 119 |
+
| `S14.sampled_qa` | 200 | chosen for the S14 human spot-check (`reports/s14_sample.tsv`) |
|
| 120 |
+
|
| 121 |
+
### Training with a BAGEL-family loader
|
| 122 |
+
|
| 123 |
+
Rows here have 2 to 7 input images: the first `num_input_images` entries of `image_list` are inputs and the rest are targets, so the loader must read `num_input_images`. The `UnifiedEditIterableDataset` of the IPT release (https://github.com/weikaih04/Imaginative-Perception-Token, `data/interleave_datasets/edit_dataset.py`) does: its `parse_row` conditions on `image_list[:num_input_images]` and trains the remaining images as targets, one after each `output_text_list` element but the last. The stock ThinkMorph loader (the same class in https://github.com/ThinkMorph/ThinkMorph) does not: it conditions on `image_list[0]` only and trains `image_list[j+1]` after `output_text_list[j]`, the answer element included, so every input after the first is trained as a generated image and every target moves one slot later per extra input. Every row here has one target, so on a row with exactly two inputs it trains `image_list[1]` (the second input view) after the plan and `image_list[2]` (the real target) after the answer; on a row with three or more inputs it trains `image_list[1]` after the plan and `image_list[2]` after the answer, both of them input views, and the real target is never trained. It raises no error. To use it, make two changes in its `parse_row`:
|
| 124 |
+
|
| 125 |
+
```
|
| 126 |
+
k = int(row.get("num_input_images", 1) or 1)
|
| 127 |
+
for im in images[:k]: # replaces the single _add_image(images[0], ...)
|
| 128 |
+
data = self._add_image(data, pil_img2rgb(Image.open(io.BytesIO(im))),
|
| 129 |
+
need_loss=False, need_vae=True, need_vit=True)
|
| 130 |
+
...
|
| 131 |
+
img_idx = idx + k # replaces img_idx = idx + 1
|
| 132 |
+
```
|
| 133 |
+
|
| 134 |
+
The stock BAGEL edit loader (ByteDance-Seed/Bagel) cannot train these rows: it never reads `output_text_list` and expects each `instruction_list` element to be a list of paraphrases.
|
| 135 |
+
|
| 136 |
+
`parquet_info.json` keys each training chunk as `<source>/<split>/<file>`, here `messytable/train/chunk_00000.parquet`, with row-group counts read from the parquet footers. The loader matches a chunk only when its key equals the path it builds, `os.path.join(data_dir, file)`, and skips a chunk with no key without a warning: a source that is alone in its group then fails with `IndexError: list index out of range`, and in a mixed group it adds no rows. Download into a directory named after the source, not after the repository:
|
| 137 |
+
|
| 138 |
+
```python
|
| 139 |
+
from huggingface_hub import snapshot_download
|
| 140 |
+
snapshot_download("yrlyrl/spatial-mmcot-messytable", repo_type="dataset", local_dir="<root>/messytable",
|
| 141 |
+
allow_patterns=["train/*", "validation/*", "parquet_info.json", "reports/known_issues/*"])
|
| 142 |
+
```
|
| 143 |
+
|
| 144 |
+
Then either run from `<root>` with `data_dir: messytable/train` and `parquet_info_path: messytable/parquet_info.json`, or rebuild the index with absolute keys and use an absolute `data_dir`:
|
| 145 |
+
|
| 146 |
+
```python
|
| 147 |
+
import json, os
|
| 148 |
+
root = "/abs/path/to/root" # the directory that holds messytable/
|
| 149 |
+
info = json.load(open(os.path.join(root, "messytable", "parquet_info.json")))
|
| 150 |
+
info = {os.path.join(root, k): v for k, v in info.items()}
|
| 151 |
+
json.dump(info, open(os.path.join(root, "messytable", "parquet_info_abs.json"), "w"))
|
| 152 |
+
# data_dir = os.path.join(root, "messytable", "train") (spelled exactly so, no trailing slash)
|
| 153 |
+
# parquet_info_path = os.path.join(root, "messytable", "parquet_info_abs.json")
|
| 154 |
+
```
|
| 155 |
+
|
| 156 |
+
The Hugging Face cache (`.../snapshots/<hash>/train/`) or a folder named `spatial-mmcot-messytable` matches no key.
|
| 157 |
+
|
| 158 |
+
`num_used_data` counts chunk files, not rows: the loader repeats this source's file list up to that number, lists every (file, row group) pair, and deals whole row groups out, floor(R / world_size) to each rank and floor(that / num_workers) to each DataLoader worker. The remainder is never read. This source has 1 training chunk file holding 12 row groups of up to 128 rows, so keep `num_used_data` large, e.g. the 128 of ThinkMorph's `interleaved_reasoning.yaml` (upstream's `example.yaml` asks for more than GPUs x workers); every row group is then read. Set to 1 and alone in its group on 8 GPUs with 4 workers, it gives every DataLoader worker an empty list, and the iterator then loops forever printing `repeat` without yielding a row. In a run that mixes sources, give each source the same multiple of its own training chunk-file count, e.g. 128 per file (128 here): the file list is repeated up to `num_used_data` entries, so a flat 128 for every source would read a two-file source's rows half as often as a one-file source's.
|
| 159 |
+
|
| 160 |
+
## How the rows were chosen
|
| 161 |
+
|
| 162 |
+
| stage | rows |
|
| 163 |
+
|---|---:|
|
| 164 |
+
| upstream rows read (train split only) | 1,880 |
|
| 165 |
+
| refused before conversion (`S0raw`) | 0 |
|
| 166 |
+
| quarantined at S4c (an automatic check could not match the read-back's conclusion to the label) | 61 |
|
| 167 |
+
| after conversion and per-row filters | 1,819 |
|
| 168 |
+
| removed by S10 (none) | 0 |
|
| 169 |
+
| removed by answer-prior balancing (S13) | 302 |
|
| 170 |
+
| **released** | **1,517** |
|
| 171 |
+
|
| 172 |
+
Every removed row has one line, with its reason, in `reports/`:
|
| 173 |
+
|
| 174 |
+
| file | step | reason (the line's `flag`, or the field shown) | rows |
|
| 175 |
+
|---|---|---|---:|
|
| 176 |
+
| `build/quarantine.jsonl` | S4c | `S4c.cot_label_conflict` | 61 |
|
| 177 |
+
| `s13_dropped.jsonl` | S13 | `step: answer` | 302 |
|
| 178 |
+
|
| 179 |
+
Every line of `s13_dropped.jsonl` has `reason: prior_downsample`; `step` names the balancing pass that removed it, and `split` is written `train` or `val` (the Hub's `validation`).
|
| 180 |
+
|
| 181 |
+
61 rows were quarantined (S4c) and are not in this release. On every one the read-back's final sentence states a total for the object the question asks about, and that total is higher than the stored label on all 61, by 1 to 14 (each line's `detail` in `build/quarantine.jsonl` gives both numbers). All of them were read: none is a phrasing mismatch. Whether the label or the read-back is wrong was not checked against every photograph (see Known issues on how the labels count).
|
| 182 |
+
|
| 183 |
+
<details><summary>Per-step counters of the conversion</summary>
|
| 184 |
+
|
| 185 |
+
1,880 upstream rows were read; `S0raw` refused 0 before a row existed and passed 1,880 to the first step. `S0` runs once more, last, on the final bytes. The reason for every refused, dropped or quarantined row is in the files above.
|
| 186 |
+
|
| 187 |
+
| step | in | out | dropped | quarantined | rejected | repaired |
|
| 188 |
+
|---|---:|---:|---:|---:|---:|---:|
|
| 189 |
+
| S4 | 1,880 | 1,880 | 0 | 0 | 0 | 0 |
|
| 190 |
+
| S4c | 1,880 | 1,819 | 0 | 61 | 0 | 0 |
|
| 191 |
+
| S5 | 1,819 | 1,819 | 0 | 0 | 0 | 0 |
|
| 192 |
+
| S8 | 1,819 | 1,819 | 0 | 0 | 0 | 0 |
|
| 193 |
+
| S9 | 1,819 | 1,819 | 0 | 0 | 0 | 0 |
|
| 194 |
+
| S0 (final structural check, after S9) | 1,819 | 1,819 | 0 | 0 | 0 | 0 |
|
| 195 |
+
|
| 196 |
+
</details>
|
| 197 |
+
|
| 198 |
+
The train/validation split keeps rows sharing a `scene_id` in `meta` on one side, and the assignment is frozen (`splits/` in the summary repository). S12 saw 1,819 rows under 191 keys. No validation input image has the content of a training input image, and none is a pixel-level near-copy of one. S12 does not record per source whether that test ran, but it skips it only for a source whose spec sets `split_leak_pixels: false`, and no spec does; over all sources it compared 21,661 candidate pairs (perceptual hash within 6 bits) pixel by pixel and found no near-copy (checked 2026-09-25).
|
| 199 |
+
|
| 200 |
+
### Answer-prior balancing (S13)
|
| 201 |
+
|
| 202 |
+
Each (task, split) group is checked separately. An answer is the answer value compared as lower-cased text without a trailing full stop, with 'farther' read as 'further' and 'nearer' as 'closer' (for multiple choice, the option text, not the letter; where the candidates are drawn in the image, as in zebra_jigsaw and zebra_tetris, the answer is the letter itself). An answer is real when it holds at least 5 rows and 2% of the group; k is the number of real answers. Answer step: the target is max(30%, 1/k) when k >= 2, and max(30%, 1/d) over the d distinct answers when k = 1; a validation group uses the larger of its own target and its task's train target. A group is cut only when k >= 1 and its most common answer holds more than the target plus 5 percentage points; every answer is then capped at one common count, chosen so that none exceeds the target, and smaller answers keep all their rows. At the answer step, a group at or below that trigger, or with no real answer (k = 0), is left as it is, so its most common answer can hold up to the target plus 5 percentage points. A task whose train group has exactly two real answers is instead cut, in every split, so that its two largest answers have equal counts, with no trigger. Rank and label steps: then, in a group where every option value of every row is a number, the rank of the correct option among the sorted values, and after it, in a group where every trained answer is an option label, the label, are each capped by the same cut-and-trigger rule on their own counts (own target, validation included): capped, never evened out, so two labels are cut only when one exceeds 55%, and then only down to 50%. These steps can also cut groups the answer step left whole, including k = 0 groups, and can raise an answer's final share above its target; the run fails if a real answer ends above the target plus 5 percentage points. A train group of at least 20 rows in which one answer holds 90% or more fails the run. PET (exact_cells_pet) instead cuts each (question type x turn direction) cell to equal counts of its two answers; a PET cell that shows only one answer is removed.
|
| 203 |
+
|
| 204 |
+
| task | split | pass | rule | rows in → out | real answers k | target | largest share, before → after | cut |
|
| 205 |
+
|---|---|---|---|---:|---:|---:|---:|---|
|
| 206 |
+
| multiview_counting | train | answer | `cap30[canon]` | 1,792 → 1,490 | 5 | 30.0% | 41.8% → 30.0% | yes |
|
| 207 |
+
| multiview_counting | validation | answer | `cap30[canon]` | 27 → 27 | 3 | 33.3% | 33.3% → 33.3% | no |
|
| 208 |
+
|
| 209 |
+
S13 removed 302 rows from this source.
|
| 210 |
+
|
| 211 |
+
### Text-only baselines
|
| 212 |
+
|
| 213 |
+
Accuracy of guessers that never see an image. For each task the released training rows are split into two fixed halves by a hash of `row_uid`; each guesser is fitted on one half and scored once on the other (one held-out half, not cross-validation; `eval rows` below). The reference is chance (the mean of 1 / number of options) where every row is multiple choice, and otherwise the eval-half accuracy of always giving the answer most common in the fit half (when a task's top answers are nearly tied, this need not be the task's most common answer; the line after the table gives that answer's validation score). Accuracies are recounted from the stored rates and `eval rows`, so they are exact. A task is flagged when a text-only guesser beats its reference by more than 0.15 (for a free-form task, a guesser other than the most common answer). A flagged task can be partly answered from the text alone; an unflagged task passed only these probes, which do not prove the text carries no answer. Report scores on every task next to this baseline.
|
| 214 |
+
|
| 215 |
+
Guessers: `keywords`: the most common answer per set of spatial words in the question; `majority`: the answer most common in the fit half; `template`: the most common answer per question wording (numbers masked, object names kept).
|
| 216 |
+
|
| 217 |
+
| task | best text-only guesser | accuracy | reference | margin | eval rows | flagged |
|
| 218 |
+
|---|---|---:|---:|---:|---:|---|
|
| 219 |
+
| multiview_counting | `template` | 0.421 | 0.298 (majority) | +0.123 | 731 | no |
|
| 220 |
+
|
| 221 |
+
Always giving the most common training answer, scored on the validation split (the constant baseline to compare validation scores with): multiview_counting: always answering `2` (30.0% of training rows) scores 0.296 (8/27).
|
| 222 |
+
|
| 223 |
+
## Spot-check (S14)
|
| 224 |
+
|
| 225 |
+
**Pending.** The S14 rows are chosen and flagged `S14.sampled_qa` in `meta` and `preview`; the human pass over them has not been signed off yet.
|
| 226 |
+
|
| 227 |
+
## Citation
|
| 228 |
+
|
| 229 |
+
Please cite MessyTable, whose photographs these are (bibtex from its repository), and credit the question and reasoning release [`leo66666/messytable`](https://huggingface.co/datasets/leo66666/messytable), whose card gives no citation:
|
| 230 |
+
|
| 231 |
+
```bibtex
|
| 232 |
+
@inproceedings{CaiZhang2020MessyTable,
|
| 233 |
+
title={MessyTable: Instance Association in Multiple Camera Views},
|
| 234 |
+
author={Zhongang Cai and Junzhe Zhang and Daxuan Ren and Cunjun Yu and Haiyu Zhao and Shuai Yi and Chai Kiat Yeo and Chen Change Loy},
|
| 235 |
+
booktitle={Proceedings of the European Conference on Computer Vision (ECCV)},
|
| 236 |
+
month={August},
|
| 237 |
+
year={2020}
|
| 238 |
+
}
|
| 239 |
+
```
|
| 240 |
+
|
| 241 |
+
## Provenance
|
| 242 |
+
|
| 243 |
+
The release files were written by our conversion code (the code repository is not public yet), `scripts/convert/export.py` at commit `4aaee1f4e946`, from build `messytable_r2`. The build was made by `scripts/convert/run_source.py` from the same repository at commit `949af62f8ab7`. S10, S12 and S13 ran before the export; `reports/export_manifest.json` pins every input the export read by SHA-1 (`build_manifest_sha1`, `s10_keep_sha1`, `s12_assignments_sha1`, `s13_balanced_keep_sha1`).
|
| 244 |
+
|
| 245 |
+
Every row removed between upstream and this release has one line, with its reason, in `reports/`: `build/dropped.jsonl` (rows refused before conversion or dropped by a conversion step); `build/quarantine.jsonl` (rows set aside by S4c because an automatic check could not match the read-back's conclusion to the label); `s10_dropped.jsonl` (duplicates removed by S10); `s10_label_conflicts.jsonl` (rows S10 withheld because another row asks the identical question, options in the same order, of the same images with a different answer); `s13_dropped.jsonl` (rows removed by answer-prior balancing). `known_issues/` lists rows with a measured problem (see Known issues); `reports/` also holds the build manifest (absolute paths cut to basenames) and counters, the S14 sample list (`s14_sample.tsv`: row_uid, task, split) and `export_manifest.json`. Part of [`yrlyrl/spatial-mmcot`](https://huggingface.co/datasets/yrlyrl/spatial-mmcot).
|
meta/train/chunk_00000.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:353cf1e66bca2755aca14c2392f2de40e648f0d1d181b6347f9e8a1750a1f8ab
|
| 3 |
+
size 96585
|
meta/validation/chunk_00000.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:bcd9b3dd0f90d7656e443b31e741a3420024da7e325a9535a86dab2807125e98
|
| 3 |
+
size 6760
|
parquet_info.json
ADDED
|
@@ -0,0 +1,10 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"messytable/train/chunk_00000.parquet": {
|
| 3 |
+
"num_row_groups": 12,
|
| 4 |
+
"num_rows": 1490
|
| 5 |
+
},
|
| 6 |
+
"messytable/validation/chunk_00000.parquet": {
|
| 7 |
+
"num_row_groups": 1,
|
| 8 |
+
"num_rows": 27
|
| 9 |
+
}
|
| 10 |
+
}
|
preview/train/chunk_00000.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:1f175ed1efe6437b717da13fcc74b9e273791d6df162232e163c2ce91a99d02f
|
| 3 |
+
size 439537067
|
preview/validation/chunk_00000.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:2c565e27cb6ec47c1d27580ef6dcbc45a8d9a3e408cd820e9659c436957b88c3
|
| 3 |
+
size 8414251
|
reports/build/counters.json
ADDED
|
@@ -0,0 +1,99 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"slug": "messytable",
|
| 3 |
+
"targets_out": 1819,
|
| 4 |
+
"steps": {
|
| 5 |
+
"S4": {
|
| 6 |
+
"step": "S4",
|
| 7 |
+
"in_": 1880,
|
| 8 |
+
"out": 1880,
|
| 9 |
+
"dropped": 0,
|
| 10 |
+
"repaired": 0,
|
| 11 |
+
"quarantined": 0,
|
| 12 |
+
"rejected": 0,
|
| 13 |
+
"seconds": 0.06309199333190918
|
| 14 |
+
},
|
| 15 |
+
"S4c": {
|
| 16 |
+
"step": "S4c",
|
| 17 |
+
"in_": 1880,
|
| 18 |
+
"out": 1819,
|
| 19 |
+
"dropped": 0,
|
| 20 |
+
"repaired": 0,
|
| 21 |
+
"quarantined": 61,
|
| 22 |
+
"rejected": 0,
|
| 23 |
+
"seconds": 0.08016562461853027
|
| 24 |
+
},
|
| 25 |
+
"S5": {
|
| 26 |
+
"step": "S5",
|
| 27 |
+
"in_": 1819,
|
| 28 |
+
"out": 1819,
|
| 29 |
+
"dropped": 0,
|
| 30 |
+
"repaired": 0,
|
| 31 |
+
"quarantined": 0,
|
| 32 |
+
"rejected": 0,
|
| 33 |
+
"seconds": 0.10321402549743652
|
| 34 |
+
},
|
| 35 |
+
"S8": {
|
| 36 |
+
"step": "S8",
|
| 37 |
+
"in_": 1819,
|
| 38 |
+
"out": 1819,
|
| 39 |
+
"dropped": 0,
|
| 40 |
+
"repaired": 0,
|
| 41 |
+
"quarantined": 0,
|
| 42 |
+
"rejected": 0,
|
| 43 |
+
"seconds": 19.548330545425415
|
| 44 |
+
},
|
| 45 |
+
"S9": {
|
| 46 |
+
"step": "S9",
|
| 47 |
+
"in_": 1819,
|
| 48 |
+
"out": 1819,
|
| 49 |
+
"dropped": 0,
|
| 50 |
+
"repaired": 0,
|
| 51 |
+
"quarantined": 0,
|
| 52 |
+
"rejected": 0,
|
| 53 |
+
"seconds": 0.318220853805542
|
| 54 |
+
},
|
| 55 |
+
"S0": {
|
| 56 |
+
"step": "S0",
|
| 57 |
+
"in_": 1819,
|
| 58 |
+
"out": 1819,
|
| 59 |
+
"dropped": 0,
|
| 60 |
+
"repaired": 0,
|
| 61 |
+
"quarantined": 0,
|
| 62 |
+
"rejected": 0,
|
| 63 |
+
"seconds": 0.7612514495849609
|
| 64 |
+
}
|
| 65 |
+
},
|
| 66 |
+
"expectations": {
|
| 67 |
+
"S4.in_": {
|
| 68 |
+
"want": 1880,
|
| 69 |
+
"got": 1880,
|
| 70 |
+
"ok": true
|
| 71 |
+
},
|
| 72 |
+
"S4.repaired": {
|
| 73 |
+
"want": 0,
|
| 74 |
+
"got": 0,
|
| 75 |
+
"ok": true
|
| 76 |
+
},
|
| 77 |
+
"S0.rejected": {
|
| 78 |
+
"want": 0,
|
| 79 |
+
"got": 0,
|
| 80 |
+
"ok": true
|
| 81 |
+
},
|
| 82 |
+
"S4c.quarantined": {
|
| 83 |
+
"want": [
|
| 84 |
+
55,
|
| 85 |
+
70
|
| 86 |
+
],
|
| 87 |
+
"got": 61,
|
| 88 |
+
"ok": true
|
| 89 |
+
},
|
| 90 |
+
"S0.out": {
|
| 91 |
+
"want": [
|
| 92 |
+
1800,
|
| 93 |
+
1880
|
| 94 |
+
],
|
| 95 |
+
"got": 1819,
|
| 96 |
+
"ok": true
|
| 97 |
+
}
|
| 98 |
+
}
|
| 99 |
+
}
|
reports/build/dropped.jsonl
ADDED
|
File without changes
|
reports/build/manifest.json
ADDED
|
@@ -0,0 +1,49 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"slug": "messytable",
|
| 3 |
+
"raw": "messytable",
|
| 4 |
+
"out": "messytable_r2",
|
| 5 |
+
"raw_rows_read": 1880,
|
| 6 |
+
"raw_unusable": 0,
|
| 7 |
+
"rows_out": 1819,
|
| 8 |
+
"targets_out": 1819,
|
| 9 |
+
"chunks": 1,
|
| 10 |
+
"row_groups": 15,
|
| 11 |
+
"seconds": 192.9,
|
| 12 |
+
"spec": {
|
| 13 |
+
"slug": "messytable",
|
| 14 |
+
"num_inputs": 7,
|
| 15 |
+
"max_k": 1,
|
| 16 |
+
"source_scene_corpus": "messytable",
|
| 17 |
+
"min_inputs": 2,
|
| 18 |
+
"license": "undeclared",
|
| 19 |
+
"license_commercial": false,
|
| 20 |
+
"replayable": false,
|
| 21 |
+
"min_image_edge": 256,
|
| 22 |
+
"copy_max_mean_diff": 2.0,
|
| 23 |
+
"copy_max_changed_frac": 0.01,
|
| 24 |
+
"allow_truncation": false,
|
| 25 |
+
"exclude_row_uids": "",
|
| 26 |
+
"exclude_row_uids_expected": 0,
|
| 27 |
+
"exclude_image_hashes": "",
|
| 28 |
+
"exclude_image_hashes_expected": 0,
|
| 29 |
+
"crop_black_bars": false,
|
| 30 |
+
"expected": {
|
| 31 |
+
"S4.in_": 1880,
|
| 32 |
+
"S4.repaired": 0,
|
| 33 |
+
"S0.rejected": 0,
|
| 34 |
+
"S4c.quarantined": [
|
| 35 |
+
55,
|
| 36 |
+
70
|
| 37 |
+
],
|
| 38 |
+
"S0.out": [
|
| 39 |
+
1800,
|
| 40 |
+
1880
|
| 41 |
+
]
|
| 42 |
+
}
|
| 43 |
+
},
|
| 44 |
+
"gates_failed": false,
|
| 45 |
+
"limit": 0,
|
| 46 |
+
"converter_git_sha": "949af62f8ab701c58da5cdeed32946242c571c83",
|
| 47 |
+
"converter_git_dirty": false,
|
| 48 |
+
"paths_redacted": "absolute build-machine paths cut to basenames for publication; build_manifest_sha1 in export_manifest.json is the original's sha1"
|
| 49 |
+
}
|
reports/build/quarantine.jsonl
ADDED
|
@@ -0,0 +1,61 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"row_uid": "4ea61052882522ca", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '2' but the label is '1'"}
|
| 2 |
+
{"row_uid": "e8508e19c2b077da", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 3 |
+
{"row_uid": "8358f091232794f3", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '3' but the label is '2'"}
|
| 4 |
+
{"row_uid": "75460d9ad0528c8f", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '2' but the label is '1'"}
|
| 5 |
+
{"row_uid": "9833c4792999ba58", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 6 |
+
{"row_uid": "6161bb2c8c4d1995", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '3' but the label is '2'"}
|
| 7 |
+
{"row_uid": "689ee435652644de", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '3' but the label is '1'"}
|
| 8 |
+
{"row_uid": "283bb8a879d9488e", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '3'"}
|
| 9 |
+
{"row_uid": "0c8f4c46e8e84712", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '2' but the label is '1'"}
|
| 10 |
+
{"row_uid": "2920f0a1bba03167", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '5' but the label is '2'"}
|
| 11 |
+
{"row_uid": "4af5c8e1380a00bc", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '3'"}
|
| 12 |
+
{"row_uid": "ac4a84f2efc1f245", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '8' but the label is '2'"}
|
| 13 |
+
{"row_uid": "646d11a1c050ea04", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '2'"}
|
| 14 |
+
{"row_uid": "86d7b52ac5da97a5", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '3' but the label is '2'"}
|
| 15 |
+
{"row_uid": "9c5f13c4450a24b6", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '3' but the label is '2'"}
|
| 16 |
+
{"row_uid": "f30d1bf6b1dee5f7", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 17 |
+
{"row_uid": "8ce99101e9964816", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 18 |
+
{"row_uid": "275227a63fad5c5a", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 19 |
+
{"row_uid": "673d1ad8bbdf0cc4", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 20 |
+
{"row_uid": "fa311fd3e0359fbd", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '2'"}
|
| 21 |
+
{"row_uid": "f7abfa003974ace8", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '2'"}
|
| 22 |
+
{"row_uid": "6291495275fa96d8", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '4'"}
|
| 23 |
+
{"row_uid": "0c42a7f6835722ee", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '5' but the label is '3'"}
|
| 24 |
+
{"row_uid": "47ad7e8d3b859f2f", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '3' but the label is '2'"}
|
| 25 |
+
{"row_uid": "96fd237c8cee31c3", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '1'"}
|
| 26 |
+
{"row_uid": "dd83d8f883237d69", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '3'"}
|
| 27 |
+
{"row_uid": "c8306aed6e2752a4", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 28 |
+
{"row_uid": "7d1ade3ef36cb8fd", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '9' but the label is '3'"}
|
| 29 |
+
{"row_uid": "5bdcda7036aca0ca", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '8' but the label is '2'"}
|
| 30 |
+
{"row_uid": "dc31d5b1cb39efe5", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '5' but the label is '2'"}
|
| 31 |
+
{"row_uid": "75668c010659040c", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '3'"}
|
| 32 |
+
{"row_uid": "07e56cea26a6f499", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '3' but the label is '1'"}
|
| 33 |
+
{"row_uid": "32ed6b8a5bc5ffe5", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '3' but the label is '2'"}
|
| 34 |
+
{"row_uid": "dc13008597d1d17e", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 35 |
+
{"row_uid": "551b7ae621313403", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '5' but the label is '2'"}
|
| 36 |
+
{"row_uid": "a2c10740af5da154", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '8' but the label is '4'"}
|
| 37 |
+
{"row_uid": "7d373bf8bc732aa9", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '2'"}
|
| 38 |
+
{"row_uid": "4816a2384df963dd", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '12' but the label is '2'"}
|
| 39 |
+
{"row_uid": "1644d4c26e48080d", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '1'"}
|
| 40 |
+
{"row_uid": "3a5906f5ba148eb0", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '10' but the label is '2'"}
|
| 41 |
+
{"row_uid": "65378e03185a1dc0", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '12' but the label is '5'"}
|
| 42 |
+
{"row_uid": "f3b8cabe8a51d858", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '12' but the label is '4'"}
|
| 43 |
+
{"row_uid": "17edb3dcb6052a3b", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '8' but the label is '4'"}
|
| 44 |
+
{"row_uid": "4ef907178906e737", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '7' but the label is '4'"}
|
| 45 |
+
{"row_uid": "a1e06c5a4f768d95", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '8' but the label is '4'"}
|
| 46 |
+
{"row_uid": "19049831769b81c8", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '8' but the label is '4'"}
|
| 47 |
+
{"row_uid": "ddf83326eb64ed72", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '2'"}
|
| 48 |
+
{"row_uid": "44729c0c93f71560", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '3'"}
|
| 49 |
+
{"row_uid": "5c4add7b9aeae51f", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 50 |
+
{"row_uid": "a731ed5b29e47dd3", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '7' but the label is '1'"}
|
| 51 |
+
{"row_uid": "85f008efe3300518", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '9' but the label is '3'"}
|
| 52 |
+
{"row_uid": "017d604bc332145e", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '13' but the label is '5'"}
|
| 53 |
+
{"row_uid": "a72c68947035897a", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '18' but the label is '4'"}
|
| 54 |
+
{"row_uid": "1c13f08936a16ba8", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '7' but the label is '3'"}
|
| 55 |
+
{"row_uid": "6c905b6b2e755f95", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '3'"}
|
| 56 |
+
{"row_uid": "b0646d36dab1bd23", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '4' but the label is '2'"}
|
| 57 |
+
{"row_uid": "370d396936cb9275", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '10' but the label is '4'"}
|
| 58 |
+
{"row_uid": "652d92d184e4a745", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '8' but the label is '4'"}
|
| 59 |
+
{"row_uid": "abcd2868ef6902ad", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '11' but the label is '4'"}
|
| 60 |
+
{"row_uid": "3978a9b602691a89", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '12' but the label is '4'"}
|
| 61 |
+
{"row_uid": "12591a52194f5bf0", "step": "S4c", "flag": "S4c.cot_label_conflict", "detail": "read-back's final sentence concludes '6' but the label is '3'"}
|
reports/export_manifest.json
ADDED
|
@@ -0,0 +1,261 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"source": "messytable",
|
| 3 |
+
"build": "messytable_r2",
|
| 4 |
+
"export_git_sha": "4aaee1f4e9469da41810aac049fa411c1834b809",
|
| 5 |
+
"export_git_dirty": false,
|
| 6 |
+
"limit": 0,
|
| 7 |
+
"converter_git_sha": "949af62f8ab701c58da5cdeed32946242c571c83",
|
| 8 |
+
"converter_git_dirty": false,
|
| 9 |
+
"build_counts": {
|
| 10 |
+
"raw_rows_read": 1880,
|
| 11 |
+
"raw_unusable": 0,
|
| 12 |
+
"rows_out": 1819
|
| 13 |
+
},
|
| 14 |
+
"build_manifest_sha1": "86277d88c96db3d53023c731ab75dc979bca4385",
|
| 15 |
+
"s10_keep_sha1": "9fd27b7ff9e2be8d9e0466795f251875210de152",
|
| 16 |
+
"s12_assignments_sha1": "d0195205cd5852e08769b99f9c72740f8ce9fcaa",
|
| 17 |
+
"s13_balanced_keep_sha1": "af73ed4ef438de882feeee30de4a25701f2a3f6a",
|
| 18 |
+
"license": "undeclared",
|
| 19 |
+
"license_commercial": false,
|
| 20 |
+
"upstream_repo": "leo66666/messytable",
|
| 21 |
+
"donor": null,
|
| 22 |
+
"visual_options": null,
|
| 23 |
+
"num_inputs": [
|
| 24 |
+
2,
|
| 25 |
+
7
|
| 26 |
+
],
|
| 27 |
+
"max_k": 1,
|
| 28 |
+
"preview_max_k": 1,
|
| 29 |
+
"rows": {
|
| 30 |
+
"train": 1490,
|
| 31 |
+
"validation": 27
|
| 32 |
+
},
|
| 33 |
+
"targets": {
|
| 34 |
+
"train": 1490,
|
| 35 |
+
"validation": 27
|
| 36 |
+
},
|
| 37 |
+
"distinct_targets_total": 1517,
|
| 38 |
+
"num_input_images": {
|
| 39 |
+
"train": [
|
| 40 |
+
2,
|
| 41 |
+
7
|
| 42 |
+
],
|
| 43 |
+
"validation": [
|
| 44 |
+
2,
|
| 45 |
+
3
|
| 46 |
+
]
|
| 47 |
+
},
|
| 48 |
+
"num_steps": {
|
| 49 |
+
"train": [
|
| 50 |
+
1,
|
| 51 |
+
1
|
| 52 |
+
],
|
| 53 |
+
"validation": [
|
| 54 |
+
1,
|
| 55 |
+
1
|
| 56 |
+
]
|
| 57 |
+
},
|
| 58 |
+
"distinct_targets": {
|
| 59 |
+
"train": 1490,
|
| 60 |
+
"validation": 27
|
| 61 |
+
},
|
| 62 |
+
"supervision_kind": {
|
| 63 |
+
"train": {
|
| 64 |
+
"full_interleaved": 1490
|
| 65 |
+
},
|
| 66 |
+
"validation": {
|
| 67 |
+
"full_interleaved": 27
|
| 68 |
+
}
|
| 69 |
+
},
|
| 70 |
+
"source_scene_corpus": {
|
| 71 |
+
"train": {
|
| 72 |
+
"messytable": 1490
|
| 73 |
+
},
|
| 74 |
+
"validation": {
|
| 75 |
+
"messytable": 27
|
| 76 |
+
}
|
| 77 |
+
},
|
| 78 |
+
"filter_flags": {
|
| 79 |
+
"train": {
|
| 80 |
+
"S5.replay_unsupported": 1490,
|
| 81 |
+
"S14.sampled_qa": 197
|
| 82 |
+
},
|
| 83 |
+
"validation": {
|
| 84 |
+
"S5.replay_unsupported": 27,
|
| 85 |
+
"S14.sampled_qa": 3
|
| 86 |
+
}
|
| 87 |
+
},
|
| 88 |
+
"answer_type": {
|
| 89 |
+
"train": {
|
| 90 |
+
"integer": 1490
|
| 91 |
+
},
|
| 92 |
+
"validation": {
|
| 93 |
+
"integer": 27
|
| 94 |
+
}
|
| 95 |
+
},
|
| 96 |
+
"answer_tag": {
|
| 97 |
+
"train": {
|
| 98 |
+
"integer": {
|
| 99 |
+
"rows": 1490,
|
| 100 |
+
"equals_answer_value": 1490
|
| 101 |
+
}
|
| 102 |
+
},
|
| 103 |
+
"validation": {
|
| 104 |
+
"integer": {
|
| 105 |
+
"rows": 27,
|
| 106 |
+
"equals_answer_value": 27
|
| 107 |
+
}
|
| 108 |
+
}
|
| 109 |
+
},
|
| 110 |
+
"options_text_rows": {
|
| 111 |
+
"train": 0,
|
| 112 |
+
"validation": 0
|
| 113 |
+
},
|
| 114 |
+
"text_checks": {
|
| 115 |
+
"train": {
|
| 116 |
+
"control_char_rows": 0,
|
| 117 |
+
"non_latin_rows": 0,
|
| 118 |
+
"empty_readback_rows": 0,
|
| 119 |
+
"placeholder_gap_rows": 0,
|
| 120 |
+
"boxed_differs_from_answer_rows": 0
|
| 121 |
+
},
|
| 122 |
+
"validation": {
|
| 123 |
+
"control_char_rows": 0,
|
| 124 |
+
"non_latin_rows": 0,
|
| 125 |
+
"empty_readback_rows": 0,
|
| 126 |
+
"placeholder_gap_rows": 0,
|
| 127 |
+
"boxed_differs_from_answer_rows": 0
|
| 128 |
+
}
|
| 129 |
+
},
|
| 130 |
+
"text_check_examples": {
|
| 131 |
+
"train": {},
|
| 132 |
+
"validation": {}
|
| 133 |
+
},
|
| 134 |
+
"row_license": {
|
| 135 |
+
"train": {
|
| 136 |
+
"undeclared": 1490
|
| 137 |
+
},
|
| 138 |
+
"validation": {
|
| 139 |
+
"undeclared": 27
|
| 140 |
+
}
|
| 141 |
+
},
|
| 142 |
+
"removed": {
|
| 143 |
+
"build/dropped.jsonl": {},
|
| 144 |
+
"build/quarantine.jsonl": {
|
| 145 |
+
"S4c": {
|
| 146 |
+
"S4c.cot_label_conflict": 61
|
| 147 |
+
}
|
| 148 |
+
},
|
| 149 |
+
"s10_dropped.jsonl": {},
|
| 150 |
+
"s10_label_conflicts.jsonl": {},
|
| 151 |
+
"s13_dropped.jsonl": {
|
| 152 |
+
"S13": {
|
| 153 |
+
"S13.answer": 302
|
| 154 |
+
}
|
| 155 |
+
}
|
| 156 |
+
},
|
| 157 |
+
"split": {
|
| 158 |
+
"policy": "scene_id",
|
| 159 |
+
"rows": 1819,
|
| 160 |
+
"keys": 191,
|
| 161 |
+
"val_rows": 27,
|
| 162 |
+
"val_keys": 3,
|
| 163 |
+
"linked_by_input": null,
|
| 164 |
+
"components": null,
|
| 165 |
+
"margin": null,
|
| 166 |
+
"pixel_near_copy_checked": null,
|
| 167 |
+
"input_leak_hits": 0,
|
| 168 |
+
"shared_target_images": 0
|
| 169 |
+
},
|
| 170 |
+
"row_uid_sha1": {
|
| 171 |
+
"train": "be1d3dc74b8a4971dcf142e460fba3e32d6375c5",
|
| 172 |
+
"validation": "6070fad95fff608a1877f80e70583c51b386bce9"
|
| 173 |
+
},
|
| 174 |
+
"tasks": {
|
| 175 |
+
"multiview_counting": {
|
| 176 |
+
"train": 1490,
|
| 177 |
+
"validation": 27
|
| 178 |
+
}
|
| 179 |
+
},
|
| 180 |
+
"answer_counts": {
|
| 181 |
+
"multiview_counting/train": {
|
| 182 |
+
"2": 447,
|
| 183 |
+
"4": 362,
|
| 184 |
+
"3": 347,
|
| 185 |
+
"1": 184,
|
| 186 |
+
"5": 119,
|
| 187 |
+
"6": 25,
|
| 188 |
+
"7": 4,
|
| 189 |
+
"8": 2
|
| 190 |
+
},
|
| 191 |
+
"multiview_counting/validation": {
|
| 192 |
+
"4": 9,
|
| 193 |
+
"2": 8,
|
| 194 |
+
"3": 5,
|
| 195 |
+
"1": 4,
|
| 196 |
+
"5": 1
|
| 197 |
+
}
|
| 198 |
+
},
|
| 199 |
+
"majority_class_rate_train": {
|
| 200 |
+
"multiview_counting": 0.3
|
| 201 |
+
},
|
| 202 |
+
"s14_sample": {
|
| 203 |
+
"rows": 200,
|
| 204 |
+
"target": 200,
|
| 205 |
+
"replay_unsupported": true
|
| 206 |
+
},
|
| 207 |
+
"s13_groups": {
|
| 208 |
+
"messytable_r2/multiview_counting/train": {
|
| 209 |
+
"rule": "cap30[canon]",
|
| 210 |
+
"rows_in": 1792,
|
| 211 |
+
"rows_out": 1490,
|
| 212 |
+
"triggered": true,
|
| 213 |
+
"exceeds_target": false,
|
| 214 |
+
"level": 447,
|
| 215 |
+
"target": 0.3,
|
| 216 |
+
"k_real": 5,
|
| 217 |
+
"distinct": 8,
|
| 218 |
+
"majority_share_in": 0.418,
|
| 219 |
+
"majority_share_out": 0.3,
|
| 220 |
+
"majority_share_final": 0.3
|
| 221 |
+
},
|
| 222 |
+
"messytable_r2/multiview_counting/val": {
|
| 223 |
+
"rule": "cap30[canon]",
|
| 224 |
+
"rows_in": 27,
|
| 225 |
+
"rows_out": 27,
|
| 226 |
+
"triggered": false,
|
| 227 |
+
"exceeds_target": false,
|
| 228 |
+
"level": 9,
|
| 229 |
+
"target": 0.3333333333333333,
|
| 230 |
+
"k_real": 3,
|
| 231 |
+
"distinct": 5,
|
| 232 |
+
"majority_share_in": 0.3333,
|
| 233 |
+
"majority_share_out": 0.3333,
|
| 234 |
+
"majority_share_final": 0.3333
|
| 235 |
+
}
|
| 236 |
+
},
|
| 237 |
+
"blind_probe": {
|
| 238 |
+
"messytable_r2/multiview_counting": {
|
| 239 |
+
"probed": true,
|
| 240 |
+
"majority": 0.2982,
|
| 241 |
+
"keywords": 0.2982,
|
| 242 |
+
"template": 0.4213,
|
| 243 |
+
"chance": null,
|
| 244 |
+
"reference": "majority",
|
| 245 |
+
"best": "template",
|
| 246 |
+
"flag": false,
|
| 247 |
+
"eval_rows": 731,
|
| 248 |
+
"sigma3": 0.0555,
|
| 249 |
+
"details": {}
|
| 250 |
+
}
|
| 251 |
+
},
|
| 252 |
+
"s13_rule": "Each (task, split) group is checked separately. An answer is the answer value compared as lower-cased text without a trailing full stop, with 'farther' read as 'further' and 'nearer' as 'closer' (for multiple choice, the option text, not the letter; where the candidates are drawn in the image, as in zebra_jigsaw and zebra_tetris, the answer is the letter itself). An answer is real when it holds at least 5 rows and 2% of the group; k is the number of real answers. Answer step: the target is max(30%, 1/k) when k >= 2, and max(30%, 1/d) over the d distinct answers when k = 1; a validation group uses the larger of its own target and its task's train target. A group is cut only when k >= 1 and its most common answer holds more than the target plus 5 percentage points; every answer is then capped at one common count, chosen so that none exceeds the target, and smaller answers keep all their rows. At the answer step, a group at or below that trigger, or with no real answer (k = 0), is left as it is, so its most common answer can hold up to the target plus 5 percentage points. A task whose train group has exactly two real answers is instead cut, in every split, so that its two largest answers have equal counts, with no trigger. Rank and label steps: then, in a group where every option value of every row is a number, the rank of the correct option among the sorted values, and after it, in a group where every trained answer is an option label, the label, are each capped by the same cut-and-trigger rule on their own counts (own target, validation included): capped, never evened out, so two labels are cut only when one exceeds 55%, and then only down to 50%. These steps can also cut groups the answer step left whole, including k = 0 groups, and can raise an answer's final share above its target; the run fails if a real answer ends above the target plus 5 percentage points. A train group of at least 20 rows in which one answer holds 90% or more fails the run. PET (exact_cells_pet) instead cuts each (question type x turn direction) cell to equal counts of its two answers; a PET cell that shows only one answer is removed.",
|
| 253 |
+
"s13_constants": {
|
| 254 |
+
"TARGET": 0.3,
|
| 255 |
+
"TRIGGER_GAP": 0.05,
|
| 256 |
+
"MIN_CLASS_ROWS": 5,
|
| 257 |
+
"MIN_CLASS_SHARE": 0.02,
|
| 258 |
+
"PURE_PRIOR": 0.9,
|
| 259 |
+
"BLIND_MARGIN": 0.15
|
| 260 |
+
}
|
| 261 |
+
}
|
reports/known_issues/readback_counts_beyond_target.tsv
ADDED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# The read-back does not count off the overhead target alone: it adds objects it says the angled input views show but the overhead crop does not (a second pineapple outside the crop, a stack seen from the side; on at least one row the added object is in no image).
|
| 2 |
+
37eff3f7b91a3eb2 train How many pineapple does this scene contain? -> 2: Cross-checking with the angled input views shows a second pineapple at the far side that is not simultaneously visible in the top-down crop, indicating two distinct pineapples in the full scene.
|
| 3 |
+
53e7468e61a0ec5b train How many clear drinking glass does this scene contain? -> 8: Cross-checking with the original angled views shows additional glasses that are stacked/overlapping and not all rims are visible from directly above; combining both angled views with the top-down layout resolves the full set to a total of 8 clear drinking g...
|
| 4 |
+
9ba7d5ebb7e52811 train What is the total number of plate in this scene? -> 4: Referring back to the angled input views, there is also an additional plate visible behind/left in the scene that is not part of those three top-down-visible groupings, bringing the total number of plates in the full scene to 4.
|
| 5 |
+
a3d46026c804984c train How many instant cup meal can you find in this scene? -> 4: I can see one cup on the upper-center plate, and another cup on the lower-left plate; cross-checking with the angled input views shows each of those positions is actually a stack of two cups (two cups at upper-right/upper-center area in the side views, and ...
|
| 6 |
+
59af2d0ca00a3795 train Please compute the number of canned beverage/food in this... -> 5: Cross-checking with the original angled views also reveals two additional cans stacked on the other cardboard block (visible as two separate can bodies in the side-angle), which are not both visible as lids in the top-down due to stacking/angle.
|
| 7 |
+
9d3d47195949f783 train How many canned beverage/food does this scene contain? -> 5: Cross-checking with the original angled views also confirms an additional can on the left side that can be easy to miss depending on angle/box occlusion.
|
| 8 |
+
3920bedde85b55f8 train How many instant cup meal are visible in this scene? -> 4: I then cross-check the wider angled view where a separate stack of instant noodle cups is visible at the edge of the table; that stack contributes one more visible instant cup meal in the scene.
|
reports/known_issues/readback_excludes_same_category.tsv
ADDED
|
@@ -0,0 +1,29 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# The question names a broad category, and the read-back reaches the (lower) label by setting aside objects of that category as a different type/brand/style ('The two orange cups are also opaque but are a different style'), by naming only the 'boxed snacks of interest', or by counting groups as one ('treating each group as one'); upstream's gt_answer appears to count one product, not the whole category.
|
| 2 |
+
40093c0df9a4d508 train Count the number of opaque cup in this scene. -> 2: e yellow striped cup near the lower-left. Only the two orange cups are opaque (the striped plastic cups), while the yellow one is a different cup type/color in this scene’s labeling. Counting just the opaque cups gives a total of 2.
|
| 3 |
+
36b8d38bf3a3396e train What is the total count for opaque cup in this scene? -> 2: cups: (1) a blue opaque mug at the upper-right and (2) a pink opaque cup near the top-center holding the banana. The two orange cups are also opaque but are a different style; however the query is specifically for “opaque cup” in this scene, which corres
|
| 4 |
+
848e7ebfd054863d train What is the count of opaque cup in this scene? -> 2: and (2) one light-blue striped tumbler near the upper-middle. The other visible vessels (two orange cups and another blue striped tumbler holding a banana) are not counted as opaque cups for this query category. Final count of opaque cups: 2.
|
| 5 |
+
d10e92abbbe74255 train Please count all sandwich in the scene. -> 2: on the right plate (the right plate also has an extra plain bread slice beside it, but that is not a sandwich). The round bun sandwich on the lower plate is a different item and not counted as a sandwich here. Counting only sandwiches in this top-down v
|
| 6 |
+
1efecc59afcb6121 train How many instances of vacuum-packed bagged snack are pres... -> 2: are the two green sealed pouches lying near the lower-left/center area. Other packages (the silver pouches and the blue/white bag) are different bag types and not counted as the queried vacuum-packed bagged snack. Counting only the green vacuum-packed snack pouches
|
| 7 |
+
20d7dacc5ee76605 train What is the final count of instant cup meal here? -> 1: ly covered by a snack bag). The smaller brown round container at the upper-left appears to be a different packaged item rather than another instant cup meal in this scene’s labeling, and no other cup-meal containers appear elsewhere on the table. Therefore, the
|
| 8 |
+
4967626ac80ae0a5 train What is the total number of vacuum-packed bagged snack in... -> 2: ear the upper-left of the cluster and one near the upper-right). Other packages (the red open-top snack bag, the blue/white pouch, cups, cans, and cartons) are not the queried vacuum-packed bagged snack type. Counting only the vacuum-packed bagged snacks gives a total of 2.
|
| 9 |
+
7df880442dad0b75 train What is the total count for boxed snack in this scene? -> 2: near the upper-left of the cluster and one blue rectangular box near the upper-middle. The other visible snack items on the right are Pocky-style boxes and are not counted as the queried boxed snack category here. Therefore, the total count of boxed snack in the scene is 2.
|
| 10 |
+
5e5b22b05fa19905 train How many instances of instant cup meal are present in thi... -> 2: with a lidded bowl-like cup stacked on/near it), and the other is at the right side on the box (a yellow cup). Other round items (like the green soup cup) are different packaging and not counted as instant cup meals. Counting only the instant cup meal instan
|
| 11 |
+
58f77ad000146905 train Give the total number of vacuum-packed bagged snack in th... -> 3: pepper, (2) a second orange packet near the center-bottom area, and (3) a third orange packet on the right box near the cabbage. The green packets appear to be different items and are not counted as vacuum-packed bagged snacks here. Total vacuum-packed b
|
| 12 |
+
917e659105a20cf0 train What is the final count of vacuum-packed bagged snack here? -> 3: green vacuum-packed snack pouches grouped on the left box. No other vacuum-packed snack packs appear elsewhere (the orange packets and the yellow chips bag are different packaging types). Therefore, the final count of vacuum-packed bagged snacks is 3.
|
| 13 |
+
98a8a62185a34161 train How many bottled beverage can you find in this scene? -> 1: ntify only one such bottle (the green-labeled drink bottle near the center cluster). The other visible containers are soda bottles and water bottles, which are not counted for this queried category. Final count of bottled beverage: 1.
|
| 14 |
+
e67b6fc5a947e432 train Can you count the boxed snack in this scene? -> 4: In the generated top-down view, the boxed snacks of interest are the blue rectangular Pocky boxes. Scanning the whole surface, I can clearly
|
| 15 |
+
7d45cb34cff3b2e9 train Find the number of instant cup meal in this scene. -> 3: : one near the bottom-center, one near the mid-right, and one near the top-right on/near the cardboard box. The central green items are instant noodle bowls (a different packaging type) and are not counted as cup meals. Counting only the cup-meal containers g
|
| 16 |
+
ddb5aa39516e9fb9 train How many boxed snack appear in this scene? -> 4: In the generated top-down view, the boxed snacks of interest are the pink rectangular “PRETZ” boxes. Scanning the whole layout: (1) one PRET
|
| 17 |
+
552c7168bc83d482 train What is the total number of boxed snack in this scene? -> 4: In the generated top-down synthesized view, the boxed snacks of interest are the blue rectangular “PRETZ” cartons. Scanning the scene systematically: (1
|
| 18 |
+
af82f5cdab0947bb train What is the scene-level count of instant cup meal? -> 3: one red-lidded cup on the right, and one purple cup near the upper-right area. Other round items (the two smaller white cups near the lower-left) appear to be different products and are not counted as instant cup meals. Counting only the instant cup meal c
|
| 19 |
+
4043d6b5e17ac81e train What is the count of bottled beverage in this scene? -> 4: oda bottles on the left stand, 2 more dark soda bottles near the upper-right area, and the other visible bottles (green tea, water, and small white drinks) are not part of the queried bottled-beverage category here. This yields a total bottled beverage count of 4.
|
| 20 |
+
681a0a7a9e50da94 train Count clear drinking glass in this scene. -> 2: near the upper-left area of the mat and one near the upper-right area). Other transparent items are different shapes (e.g., taller cylindrical glasses) and are not counted as the queried clear drinking glass. Final count: 2.
|
| 21 |
+
95b752007185ac5c train How many opaque cup are visible in this scene? -> 2: near the upper center and one light gray mug near the lower center-right. The other open vessels are colored plastic cups (not counted here as “opaque cup” in this scene’s labeling), and the remaining clear vessels are transparent glasses. Therefore, the numbe
|
| 22 |
+
976de70a4c9c5186 train Please compute the number of paper cup milk tea in this s... -> 3: ). Scanning the entire table, I see three orange-lidded milk tea cups grouped around the center-right/center area, while the purple-lidded cups correspond to a different drink type and are not counted as paper cup milk tea. Therefore, the total number of
|
| 23 |
+
4becf44125ea997c train How many boxed snack does this scene contain? -> 4: In the generated top-down view, the boxed snacks of interest are the larger rectangular carton-style snack boxes (the pink “PRETZ” cartons).
|
| 24 |
+
7438e1c7bf40505b train How many instances of plate are present in this scene? -> 2: is clear: I can see one large round plate holding stacked bread slices and one smaller round plate holding two bagels. Other visible items are bowls/plates of different types, but for the queried category ‘plate’ there are exactly these two instances. F
|
| 25 |
+
b9f6818180df44e1 train Please count all boxed snack in the scene. -> 4: oxed snacks stacked/overlapped near the right side by the cans and black box. The pink-white packages at the upper-left are a different boxed item type and are not counted as the queried boxed snack category here. Total boxed snacks counted: 4.
|
| 26 |
+
20ad13ee7c219123 train Give the total number of canned beverage/food in this scene. -> 4: tops there gives 4 cans. Other visible containers (plastic water bottles, cola bottles, and the orange soda can) are not counted because they are bottles or a different container type than the specified canned food/beverage category. Therefore, the total num
|
| 27 |
+
3679ee397a0860d4 train Count canned beverage/food in this scene. -> 5: , (3) three green beverage cans on the small center box, (4) four blue beverage cans on the right box, and (5) a small cluster of three cans at the lower-left. Treating each group as one distinct canned-beverage/food instance in the scene yields a total count
|
| 28 |
+
483b42daf677e4d9 train Please provide the number of boxed snack in this scene. -> 4: nacks; (2) on the right, the red packages appear grouped around the cardboard box but are not the queried pink boxed snack type. Counting only the boxed snacks of interest visible as individual boxes in the scene gives a total of 4.
|
| 29 |
+
2d308f1341ba5d96 validation How many bottled beverage can you find in this scene? -> 4: er by the pineapple, one near the center-left, one near the lower-center, and one near the center-right beside the black platform. The red-capped bottles are a different type and the green containers are cans, so they are excluded. Counting only bottled
|
reports/known_issues/summary.json
ADDED
|
@@ -0,0 +1,20 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"readback_counts_beyond_target": {
|
| 3 |
+
"description": "The read-back does not count off the overhead target alone: it adds objects it says the angled input views show but the overhead crop does not (a second pineapple outside the crop, a stack seen from the side; on at least one row the added object is in no image).",
|
| 4 |
+
"rows": 7,
|
| 5 |
+
"train": 7,
|
| 6 |
+
"validation": 0,
|
| 7 |
+
"file": "known_issues/readback_counts_beyond_target.tsv",
|
| 8 |
+
"detector": "a read-back sentence mentions the angled/input/original/side views and adds something (additional, extra, another, one more, a second, bringing the total, not visible in the top-down view, outside the crop, a stack of two) with no negation (no/not/without/any/rather than) in the 30 characters before it, not in a clause whose verb is 'confirm(s)' unless it confirms 'an additional/another ...', and not 'appear(s) as additional'",
|
| 9 |
+
"measured": "2026-09-25; known_issues.py sha1 655e5307; export 4aaee1f rows be1d3dc7/6070fad9"
|
| 10 |
+
},
|
| 11 |
+
"readback_excludes_same_category": {
|
| 12 |
+
"description": "The question names a broad category, and the read-back reaches the (lower) label by setting aside objects of that category as a different type/brand/style ('The two orange cups are also opaque but are a different style'), by naming only the 'boxed snacks of interest', or by counting groups as one ('treating each group as one'); upstream's gt_answer appears to count one product, not the whole category.",
|
| 13 |
+
"rows": 28,
|
| 14 |
+
"train": 27,
|
| 15 |
+
"validation": 1,
|
| 16 |
+
"file": "known_issues/readback_excludes_same_category.tsv",
|
| 17 |
+
"detector": "the read-back says 'of interest', 'treating each group' or 'this scene's labeling', or it has an exclusion phrase ('different type|kind|brand|style|product|item|drink|packaging|container', 'not counted / not the queried ... category|type', 'specified') whose subject (from the sentence start, or from an unclosed '(' that opens its own clause, to the phrase) names an object of the queried category (its non-generic words, or their container noun: bottle, can, box/carton, bag/pouch/packet, cup/mug/tumbler)",
|
| 18 |
+
"measured": "2026-09-25; known_issues.py sha1 655e5307; export 4aaee1f rows be1d3dc7/6070fad9"
|
| 19 |
+
}
|
| 20 |
+
}
|
reports/s10_dropped.jsonl
ADDED
|
File without changes
|
reports/s10_label_conflicts.jsonl
ADDED
|
File without changes
|
reports/s13_dropped.jsonl
ADDED
|
@@ -0,0 +1,302 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"build": "messytable_r2", "row_uid": "736aaf5f474afd55", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 2 |
+
{"build": "messytable_r2", "row_uid": "fc5dcc5229bf9ec3", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 3 |
+
{"build": "messytable_r2", "row_uid": "8019a6e3779b2344", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 4 |
+
{"build": "messytable_r2", "row_uid": "015ea04ed9fb6b11", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 5 |
+
{"build": "messytable_r2", "row_uid": "984887116ea4be6e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 6 |
+
{"build": "messytable_r2", "row_uid": "410b0e767ffdf3dc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 7 |
+
{"build": "messytable_r2", "row_uid": "2531c8328ab93a80", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 8 |
+
{"build": "messytable_r2", "row_uid": "44292464f4ac3376", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 9 |
+
{"build": "messytable_r2", "row_uid": "9af3d0b22eb17869", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 10 |
+
{"build": "messytable_r2", "row_uid": "c2d362f1bb283515", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 11 |
+
{"build": "messytable_r2", "row_uid": "917ccd5823c50b4c", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 12 |
+
{"build": "messytable_r2", "row_uid": "707cf82a3da77ce9", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 13 |
+
{"build": "messytable_r2", "row_uid": "8ad49c2861c68976", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 14 |
+
{"build": "messytable_r2", "row_uid": "6ac5d2690a2c9250", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 15 |
+
{"build": "messytable_r2", "row_uid": "64d4e3e5950dfdba", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 16 |
+
{"build": "messytable_r2", "row_uid": "f4a4778fd99edd5b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 17 |
+
{"build": "messytable_r2", "row_uid": "a63f2dea3e3adf54", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 18 |
+
{"build": "messytable_r2", "row_uid": "edd1922f34fc04ca", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 19 |
+
{"build": "messytable_r2", "row_uid": "bdb049904aefd06f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 20 |
+
{"build": "messytable_r2", "row_uid": "ea3c2663e137d5f8", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 21 |
+
{"build": "messytable_r2", "row_uid": "1c98aa67362c255f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 22 |
+
{"build": "messytable_r2", "row_uid": "bf98e6b31fff3f48", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 23 |
+
{"build": "messytable_r2", "row_uid": "a128de56a7547a6e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 24 |
+
{"build": "messytable_r2", "row_uid": "15d03acdaf173af7", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 25 |
+
{"build": "messytable_r2", "row_uid": "63bb1b93e4b0a3cc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 26 |
+
{"build": "messytable_r2", "row_uid": "04fb9edeef48c987", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 27 |
+
{"build": "messytable_r2", "row_uid": "f6c751ef9f68cd8a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 28 |
+
{"build": "messytable_r2", "row_uid": "a047892b0b810069", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 29 |
+
{"build": "messytable_r2", "row_uid": "9dc9b715857bc883", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 30 |
+
{"build": "messytable_r2", "row_uid": "e0e4a0b19e20685e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 31 |
+
{"build": "messytable_r2", "row_uid": "8951dcf043b7ccff", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 32 |
+
{"build": "messytable_r2", "row_uid": "b0c69daa7b48d64f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 33 |
+
{"build": "messytable_r2", "row_uid": "4f9b628018a5fe10", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 34 |
+
{"build": "messytable_r2", "row_uid": "86e6dbe0c38d02e5", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 35 |
+
{"build": "messytable_r2", "row_uid": "097152fb44febd0f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 36 |
+
{"build": "messytable_r2", "row_uid": "898efb66d1369ba0", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 37 |
+
{"build": "messytable_r2", "row_uid": "b9598c9b447338da", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 38 |
+
{"build": "messytable_r2", "row_uid": "f6c558fcbebbe91b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 39 |
+
{"build": "messytable_r2", "row_uid": "5996bc492a73ec5a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 40 |
+
{"build": "messytable_r2", "row_uid": "af9630286ae9409e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 41 |
+
{"build": "messytable_r2", "row_uid": "5b9818995baa2107", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 42 |
+
{"build": "messytable_r2", "row_uid": "f49cb759c537fd16", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 43 |
+
{"build": "messytable_r2", "row_uid": "b8b437b2aa19ecb8", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 44 |
+
{"build": "messytable_r2", "row_uid": "3c368322d6f96d6f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 45 |
+
{"build": "messytable_r2", "row_uid": "7f9de929da192d63", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 46 |
+
{"build": "messytable_r2", "row_uid": "cc1797fc9192f8fe", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 47 |
+
{"build": "messytable_r2", "row_uid": "02815c10ac46f41f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 48 |
+
{"build": "messytable_r2", "row_uid": "b83e7884194d5bd7", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 49 |
+
{"build": "messytable_r2", "row_uid": "85f50da98cedee42", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 50 |
+
{"build": "messytable_r2", "row_uid": "ac033d21630cc33a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 51 |
+
{"build": "messytable_r2", "row_uid": "0977125e5ce61aca", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 52 |
+
{"build": "messytable_r2", "row_uid": "6a221a734588e22b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 53 |
+
{"build": "messytable_r2", "row_uid": "4eca16b87efcf3dd", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 54 |
+
{"build": "messytable_r2", "row_uid": "c1616d0b7deb62dd", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 55 |
+
{"build": "messytable_r2", "row_uid": "f9ce6026b6e74f39", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 56 |
+
{"build": "messytable_r2", "row_uid": "2ef0531bf9767d6d", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 57 |
+
{"build": "messytable_r2", "row_uid": "fb805f5edbe62ae1", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 58 |
+
{"build": "messytable_r2", "row_uid": "bedb3858ad9cd4cc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 59 |
+
{"build": "messytable_r2", "row_uid": "3f072387c924b49b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 60 |
+
{"build": "messytable_r2", "row_uid": "7b34e6bc79f5acea", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 61 |
+
{"build": "messytable_r2", "row_uid": "5a190b071625e5be", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 62 |
+
{"build": "messytable_r2", "row_uid": "ef8d7fa3ad31ab6a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 63 |
+
{"build": "messytable_r2", "row_uid": "70f2d2cc0216a096", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 64 |
+
{"build": "messytable_r2", "row_uid": "50cfaee158a45698", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 65 |
+
{"build": "messytable_r2", "row_uid": "a2d6d1fa8f2d50ff", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 66 |
+
{"build": "messytable_r2", "row_uid": "8ad54818ce256df6", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 67 |
+
{"build": "messytable_r2", "row_uid": "63a0603e44bba64a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 68 |
+
{"build": "messytable_r2", "row_uid": "a4a6bb35c03b0f28", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 69 |
+
{"build": "messytable_r2", "row_uid": "265c5cb1e8bdce95", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 70 |
+
{"build": "messytable_r2", "row_uid": "4ecf681315e27b95", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 71 |
+
{"build": "messytable_r2", "row_uid": "c33c7ba6f5128b4b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 72 |
+
{"build": "messytable_r2", "row_uid": "87119b747b8e181e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 73 |
+
{"build": "messytable_r2", "row_uid": "fe7369fed05dd7f0", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 74 |
+
{"build": "messytable_r2", "row_uid": "ee038d39dcdf13bb", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 75 |
+
{"build": "messytable_r2", "row_uid": "2e78bbde1b1b723b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 76 |
+
{"build": "messytable_r2", "row_uid": "7b16ed605386ca9b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 77 |
+
{"build": "messytable_r2", "row_uid": "8cb990b1c59f085e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 78 |
+
{"build": "messytable_r2", "row_uid": "d73d29097c419592", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 79 |
+
{"build": "messytable_r2", "row_uid": "b303c08cdf2a567e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 80 |
+
{"build": "messytable_r2", "row_uid": "d918a4448826f9c4", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 81 |
+
{"build": "messytable_r2", "row_uid": "9eb3173ae3ed9b71", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 82 |
+
{"build": "messytable_r2", "row_uid": "f5ed65b8598e7f32", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 83 |
+
{"build": "messytable_r2", "row_uid": "577bc800b4395b7b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 84 |
+
{"build": "messytable_r2", "row_uid": "33fde0dd40535fd4", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 85 |
+
{"build": "messytable_r2", "row_uid": "69fb905bd448b9fa", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 86 |
+
{"build": "messytable_r2", "row_uid": "9262ae8d68f81e4b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 87 |
+
{"build": "messytable_r2", "row_uid": "557d35d33c924734", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 88 |
+
{"build": "messytable_r2", "row_uid": "7672be0f5ece42b1", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 89 |
+
{"build": "messytable_r2", "row_uid": "93505b4b3d2d5e6a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 90 |
+
{"build": "messytable_r2", "row_uid": "87ac56384ecd9c2a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 91 |
+
{"build": "messytable_r2", "row_uid": "dee5a12a74c3e314", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 92 |
+
{"build": "messytable_r2", "row_uid": "250ca0802751a942", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 93 |
+
{"build": "messytable_r2", "row_uid": "62fc151649bd2a05", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 94 |
+
{"build": "messytable_r2", "row_uid": "0ad7b19467fb0015", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 95 |
+
{"build": "messytable_r2", "row_uid": "36c5152d80780777", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 96 |
+
{"build": "messytable_r2", "row_uid": "b5624b14d9678718", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 97 |
+
{"build": "messytable_r2", "row_uid": "0bdc9f5013ee6592", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 98 |
+
{"build": "messytable_r2", "row_uid": "6a67399dcc62d7fb", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 99 |
+
{"build": "messytable_r2", "row_uid": "8a9e5442940e5c37", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 100 |
+
{"build": "messytable_r2", "row_uid": "ee9570492d4b8844", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 101 |
+
{"build": "messytable_r2", "row_uid": "6e423383a51bf910", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 102 |
+
{"build": "messytable_r2", "row_uid": "1e738f873f2e38f8", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 103 |
+
{"build": "messytable_r2", "row_uid": "97f31be43b2c0b11", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 104 |
+
{"build": "messytable_r2", "row_uid": "c3ad3fc4c1c6e349", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 105 |
+
{"build": "messytable_r2", "row_uid": "228b1ae0d4654010", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 106 |
+
{"build": "messytable_r2", "row_uid": "af390bdea7192741", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 107 |
+
{"build": "messytable_r2", "row_uid": "3650afb23c8487be", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 108 |
+
{"build": "messytable_r2", "row_uid": "2b05b778ddb4f9df", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 109 |
+
{"build": "messytable_r2", "row_uid": "1c16080bb0c29dcc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 110 |
+
{"build": "messytable_r2", "row_uid": "683ee9b4bf43901a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 111 |
+
{"build": "messytable_r2", "row_uid": "0e700bd9c4171971", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 112 |
+
{"build": "messytable_r2", "row_uid": "b3585df415763728", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 113 |
+
{"build": "messytable_r2", "row_uid": "76792e4b54e60a47", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 114 |
+
{"build": "messytable_r2", "row_uid": "c11bba068ac1fea4", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 115 |
+
{"build": "messytable_r2", "row_uid": "37c73545019c9218", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 116 |
+
{"build": "messytable_r2", "row_uid": "797ec0a13f1fef91", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 117 |
+
{"build": "messytable_r2", "row_uid": "f5d06aa3039d20c7", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 118 |
+
{"build": "messytable_r2", "row_uid": "cf170cead7480b91", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 119 |
+
{"build": "messytable_r2", "row_uid": "5822c5f5f422f411", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 120 |
+
{"build": "messytable_r2", "row_uid": "3b973982ae23c1db", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 121 |
+
{"build": "messytable_r2", "row_uid": "7a4282543487b191", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 122 |
+
{"build": "messytable_r2", "row_uid": "be66b51b4a9f2db3", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 123 |
+
{"build": "messytable_r2", "row_uid": "535f29cc41e4b3e2", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 124 |
+
{"build": "messytable_r2", "row_uid": "4bb59f698d7cf8e9", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 125 |
+
{"build": "messytable_r2", "row_uid": "1b98d913dbf85157", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 126 |
+
{"build": "messytable_r2", "row_uid": "18c96f90074a243a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 127 |
+
{"build": "messytable_r2", "row_uid": "80b50faa06c802b4", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 128 |
+
{"build": "messytable_r2", "row_uid": "846e176c96abf963", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 129 |
+
{"build": "messytable_r2", "row_uid": "791b31c4977e847b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 130 |
+
{"build": "messytable_r2", "row_uid": "d83f2b274807ca90", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 131 |
+
{"build": "messytable_r2", "row_uid": "6ed2502723310c21", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 132 |
+
{"build": "messytable_r2", "row_uid": "3ea33ad841a3b55a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 133 |
+
{"build": "messytable_r2", "row_uid": "67a8319c980a4e54", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 134 |
+
{"build": "messytable_r2", "row_uid": "0136bdf823c00ce1", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 135 |
+
{"build": "messytable_r2", "row_uid": "a091b05e64c1b406", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 136 |
+
{"build": "messytable_r2", "row_uid": "096cc359dc9fe081", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 137 |
+
{"build": "messytable_r2", "row_uid": "cb8b0c487b405d77", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 138 |
+
{"build": "messytable_r2", "row_uid": "04aabef7b35d5cc0", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 139 |
+
{"build": "messytable_r2", "row_uid": "3b6866d2e17186a8", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 140 |
+
{"build": "messytable_r2", "row_uid": "887e133054527729", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 141 |
+
{"build": "messytable_r2", "row_uid": "27e3a798786f2a5b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 142 |
+
{"build": "messytable_r2", "row_uid": "fc0386aaef692f65", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 143 |
+
{"build": "messytable_r2", "row_uid": "392f44956f313a2d", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 144 |
+
{"build": "messytable_r2", "row_uid": "4f2de79c55d9691c", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 145 |
+
{"build": "messytable_r2", "row_uid": "96d946b89ea12c4e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 146 |
+
{"build": "messytable_r2", "row_uid": "e921740be0ea4172", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 147 |
+
{"build": "messytable_r2", "row_uid": "8c0ee0abe70ff291", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 148 |
+
{"build": "messytable_r2", "row_uid": "3018b0ac8b1053ba", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 149 |
+
{"build": "messytable_r2", "row_uid": "fa65c28cba464d08", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 150 |
+
{"build": "messytable_r2", "row_uid": "79496f0902ce786b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 151 |
+
{"build": "messytable_r2", "row_uid": "6d3b63dcf820c89d", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 152 |
+
{"build": "messytable_r2", "row_uid": "e08f2cbdac46fb7c", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 153 |
+
{"build": "messytable_r2", "row_uid": "ca00407f9e7ef8b0", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 154 |
+
{"build": "messytable_r2", "row_uid": "351e598711c32d13", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 155 |
+
{"build": "messytable_r2", "row_uid": "b38914b7c71242bf", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 156 |
+
{"build": "messytable_r2", "row_uid": "149e859e1b5709f7", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 157 |
+
{"build": "messytable_r2", "row_uid": "00f3dab638a7beb6", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 158 |
+
{"build": "messytable_r2", "row_uid": "9e0dd98e77e80572", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 159 |
+
{"build": "messytable_r2", "row_uid": "2f5a387b9bd5062c", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 160 |
+
{"build": "messytable_r2", "row_uid": "45c295d165381148", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 161 |
+
{"build": "messytable_r2", "row_uid": "b6a739201810d467", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 162 |
+
{"build": "messytable_r2", "row_uid": "c351d9b571935d17", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 163 |
+
{"build": "messytable_r2", "row_uid": "693457671515cecc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 164 |
+
{"build": "messytable_r2", "row_uid": "1c7c2ed989f0c245", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 165 |
+
{"build": "messytable_r2", "row_uid": "96bb2c51e24964fc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 166 |
+
{"build": "messytable_r2", "row_uid": "948c2218547eb982", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 167 |
+
{"build": "messytable_r2", "row_uid": "666cde9689d8af46", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 168 |
+
{"build": "messytable_r2", "row_uid": "6a80aee0dfa7cf1e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 169 |
+
{"build": "messytable_r2", "row_uid": "08196b3f554f80d2", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 170 |
+
{"build": "messytable_r2", "row_uid": "d67ff13b39191a22", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 171 |
+
{"build": "messytable_r2", "row_uid": "e0dc18b5a00da6da", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 172 |
+
{"build": "messytable_r2", "row_uid": "6a216f6f8f50cb90", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 173 |
+
{"build": "messytable_r2", "row_uid": "eab78863824c1359", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 174 |
+
{"build": "messytable_r2", "row_uid": "e26fce06470f7193", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 175 |
+
{"build": "messytable_r2", "row_uid": "7fd312cbc8ef5728", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 176 |
+
{"build": "messytable_r2", "row_uid": "7e52c06c7d40e0b4", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 177 |
+
{"build": "messytable_r2", "row_uid": "4b7491d749d866c4", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 178 |
+
{"build": "messytable_r2", "row_uid": "2595d0d96e609c64", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 179 |
+
{"build": "messytable_r2", "row_uid": "0b8de84059ece6c4", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 180 |
+
{"build": "messytable_r2", "row_uid": "60cf44da95a3f763", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 181 |
+
{"build": "messytable_r2", "row_uid": "008ade289233b8f3", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 182 |
+
{"build": "messytable_r2", "row_uid": "31f0163492f95742", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 183 |
+
{"build": "messytable_r2", "row_uid": "3b3045b6b70f08e7", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 184 |
+
{"build": "messytable_r2", "row_uid": "c99684706cdf5238", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 185 |
+
{"build": "messytable_r2", "row_uid": "d25d99114c313236", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 186 |
+
{"build": "messytable_r2", "row_uid": "e4d9f927c2b32af3", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 187 |
+
{"build": "messytable_r2", "row_uid": "7998e07aa75b767b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 188 |
+
{"build": "messytable_r2", "row_uid": "9d1ec0b7ec05047f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 189 |
+
{"build": "messytable_r2", "row_uid": "36bde717042a15ad", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 190 |
+
{"build": "messytable_r2", "row_uid": "c0f8c4ff3cc647b3", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 191 |
+
{"build": "messytable_r2", "row_uid": "faa5feab066dae93", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 192 |
+
{"build": "messytable_r2", "row_uid": "3833df81f73b6cb1", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 193 |
+
{"build": "messytable_r2", "row_uid": "39d6ad18eb63afd1", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 194 |
+
{"build": "messytable_r2", "row_uid": "6221173753081a64", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 195 |
+
{"build": "messytable_r2", "row_uid": "fc07647e7c9cf74b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 196 |
+
{"build": "messytable_r2", "row_uid": "3f1d20dc35324e36", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 197 |
+
{"build": "messytable_r2", "row_uid": "fbbf30a779281f9b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 198 |
+
{"build": "messytable_r2", "row_uid": "8d4f39200136d2dd", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 199 |
+
{"build": "messytable_r2", "row_uid": "270cd3dec59ff8f3", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 200 |
+
{"build": "messytable_r2", "row_uid": "212954bc394d32dc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 201 |
+
{"build": "messytable_r2", "row_uid": "6e17f462d07199a0", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 202 |
+
{"build": "messytable_r2", "row_uid": "50b0448d1fe1e90c", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 203 |
+
{"build": "messytable_r2", "row_uid": "00e59a2995938dbd", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 204 |
+
{"build": "messytable_r2", "row_uid": "3a3097101988373f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 205 |
+
{"build": "messytable_r2", "row_uid": "8c9006694fc46127", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 206 |
+
{"build": "messytable_r2", "row_uid": "10822713337de9c6", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 207 |
+
{"build": "messytable_r2", "row_uid": "2ab3404bc34d868d", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 208 |
+
{"build": "messytable_r2", "row_uid": "0848422c90ab0dd0", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 209 |
+
{"build": "messytable_r2", "row_uid": "d60158cc38b30a90", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 210 |
+
{"build": "messytable_r2", "row_uid": "f3319227a1158a31", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 211 |
+
{"build": "messytable_r2", "row_uid": "fa4c6ed35b590982", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 212 |
+
{"build": "messytable_r2", "row_uid": "0d0f868c07fad581", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 213 |
+
{"build": "messytable_r2", "row_uid": "a477cc5188ed3d91", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 214 |
+
{"build": "messytable_r2", "row_uid": "ce04c703cc3b0d3c", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 215 |
+
{"build": "messytable_r2", "row_uid": "ede4bc9b6f64a7e7", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 216 |
+
{"build": "messytable_r2", "row_uid": "91bebebd3e8444bc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 217 |
+
{"build": "messytable_r2", "row_uid": "817eb2cf57953bff", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 218 |
+
{"build": "messytable_r2", "row_uid": "3b90b8316ca78217", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 219 |
+
{"build": "messytable_r2", "row_uid": "49adfef71ffc81d1", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 220 |
+
{"build": "messytable_r2", "row_uid": "5ac905f2bbe8e321", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 221 |
+
{"build": "messytable_r2", "row_uid": "86ecfb3fe5941d08", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 222 |
+
{"build": "messytable_r2", "row_uid": "a437955ac7a677a2", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 223 |
+
{"build": "messytable_r2", "row_uid": "12fea2c41777abe9", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 224 |
+
{"build": "messytable_r2", "row_uid": "aea977aa48f1011b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 225 |
+
{"build": "messytable_r2", "row_uid": "e715e81769d55e18", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 226 |
+
{"build": "messytable_r2", "row_uid": "6af37af45dcf9741", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 227 |
+
{"build": "messytable_r2", "row_uid": "393119af89a20217", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 228 |
+
{"build": "messytable_r2", "row_uid": "bb60dc81f3929d9f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 229 |
+
{"build": "messytable_r2", "row_uid": "888e6cc65d0bc98b", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 230 |
+
{"build": "messytable_r2", "row_uid": "61d0bf22025b204c", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 231 |
+
{"build": "messytable_r2", "row_uid": "e4c3a5086f638797", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 232 |
+
{"build": "messytable_r2", "row_uid": "8e1d2009e76184a9", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 233 |
+
{"build": "messytable_r2", "row_uid": "0f250093df06f34c", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 234 |
+
{"build": "messytable_r2", "row_uid": "66ec39b24ee0fd3e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 235 |
+
{"build": "messytable_r2", "row_uid": "8415caaf8e9fdc01", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 236 |
+
{"build": "messytable_r2", "row_uid": "07db7b75a04268f5", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 237 |
+
{"build": "messytable_r2", "row_uid": "ea4ae66311321bd8", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 238 |
+
{"build": "messytable_r2", "row_uid": "e06cfd40119fbd48", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 239 |
+
{"build": "messytable_r2", "row_uid": "096bf5b189b46926", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 240 |
+
{"build": "messytable_r2", "row_uid": "f18d2783dedcb3f1", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 241 |
+
{"build": "messytable_r2", "row_uid": "fa996bb4c3cc7dc5", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 242 |
+
{"build": "messytable_r2", "row_uid": "ead2334da86d50be", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 243 |
+
{"build": "messytable_r2", "row_uid": "2dcc59c4825cbe27", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 244 |
+
{"build": "messytable_r2", "row_uid": "7ae80f4cbab55ea0", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 245 |
+
{"build": "messytable_r2", "row_uid": "e5ff8067193e60ff", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 246 |
+
{"build": "messytable_r2", "row_uid": "8d11bd243ab4f3b1", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 247 |
+
{"build": "messytable_r2", "row_uid": "f0d38ac5af3d6fce", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 248 |
+
{"build": "messytable_r2", "row_uid": "fdc14541f61bd060", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 249 |
+
{"build": "messytable_r2", "row_uid": "7fa4e9ea9ca65ace", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 250 |
+
{"build": "messytable_r2", "row_uid": "4b97dbc68eeff0f9", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 251 |
+
{"build": "messytable_r2", "row_uid": "af310a74d71332d5", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 252 |
+
{"build": "messytable_r2", "row_uid": "54879fabe4cddfd3", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 253 |
+
{"build": "messytable_r2", "row_uid": "b48cb7020cc581f4", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 254 |
+
{"build": "messytable_r2", "row_uid": "29f111a9ce3ca6f8", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 255 |
+
{"build": "messytable_r2", "row_uid": "ee60c3a62763afbc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 256 |
+
{"build": "messytable_r2", "row_uid": "7e9f9879f7cd07e0", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 257 |
+
{"build": "messytable_r2", "row_uid": "72c27adf8089f9f5", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 258 |
+
{"build": "messytable_r2", "row_uid": "3d52ce709dfcabbb", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 259 |
+
{"build": "messytable_r2", "row_uid": "3e8bcf3f5b9d57f6", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 260 |
+
{"build": "messytable_r2", "row_uid": "d158974d95871e61", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 261 |
+
{"build": "messytable_r2", "row_uid": "ae649f7d3e6fbe7a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 262 |
+
{"build": "messytable_r2", "row_uid": "e6e9b30f01ad25a7", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 263 |
+
{"build": "messytable_r2", "row_uid": "808408c64f5d62d6", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 264 |
+
{"build": "messytable_r2", "row_uid": "8aa029178daf2fcb", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 265 |
+
{"build": "messytable_r2", "row_uid": "603c6da708e3acd3", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 266 |
+
{"build": "messytable_r2", "row_uid": "b74e51291074c7e6", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 267 |
+
{"build": "messytable_r2", "row_uid": "dae4f7c9a00aa700", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 268 |
+
{"build": "messytable_r2", "row_uid": "83a861a381cdb6ad", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 269 |
+
{"build": "messytable_r2", "row_uid": "44e66a93de33611d", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 270 |
+
{"build": "messytable_r2", "row_uid": "de06261236287272", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 271 |
+
{"build": "messytable_r2", "row_uid": "cd1effda714cfebc", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 272 |
+
{"build": "messytable_r2", "row_uid": "afe90f1226fa38e1", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 273 |
+
{"build": "messytable_r2", "row_uid": "323be2d173f48df3", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 274 |
+
{"build": "messytable_r2", "row_uid": "aa3c5987344d9d0d", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 275 |
+
{"build": "messytable_r2", "row_uid": "b1715da309e32d9f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 276 |
+
{"build": "messytable_r2", "row_uid": "762966c0432f8840", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 277 |
+
{"build": "messytable_r2", "row_uid": "17afe104b3c05e11", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 278 |
+
{"build": "messytable_r2", "row_uid": "73db6441976ceaf6", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 279 |
+
{"build": "messytable_r2", "row_uid": "35c5ac0851bdd88f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 280 |
+
{"build": "messytable_r2", "row_uid": "e7a590aec31c3b05", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 281 |
+
{"build": "messytable_r2", "row_uid": "c699d684a210a68f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 282 |
+
{"build": "messytable_r2", "row_uid": "228f341a2451affe", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 283 |
+
{"build": "messytable_r2", "row_uid": "2c21b16d917a185d", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 284 |
+
{"build": "messytable_r2", "row_uid": "61cac9baa2d84adb", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 285 |
+
{"build": "messytable_r2", "row_uid": "619f6da83c3748de", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 286 |
+
{"build": "messytable_r2", "row_uid": "cd9af4438c94a75d", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 287 |
+
{"build": "messytable_r2", "row_uid": "42fc95b1a74b3d01", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 288 |
+
{"build": "messytable_r2", "row_uid": "fef816290787c7cf", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 289 |
+
{"build": "messytable_r2", "row_uid": "279cd3039e95a312", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 290 |
+
{"build": "messytable_r2", "row_uid": "3fb9abf2bedea3d7", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 291 |
+
{"build": "messytable_r2", "row_uid": "b8bba55ea970ec52", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 292 |
+
{"build": "messytable_r2", "row_uid": "f062ab9d48e424f6", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 293 |
+
{"build": "messytable_r2", "row_uid": "fcc3b81af62ed623", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 294 |
+
{"build": "messytable_r2", "row_uid": "4daba84edf71435a", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 295 |
+
{"build": "messytable_r2", "row_uid": "267fb907a4d4b23e", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 296 |
+
{"build": "messytable_r2", "row_uid": "dbec4e14e8fe1d41", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 297 |
+
{"build": "messytable_r2", "row_uid": "391b25cef5d27250", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 298 |
+
{"build": "messytable_r2", "row_uid": "8b24d1742e3e2586", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 299 |
+
{"build": "messytable_r2", "row_uid": "fa225babe4ccad39", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 300 |
+
{"build": "messytable_r2", "row_uid": "a89d919f1fe6aa5c", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 301 |
+
{"build": "messytable_r2", "row_uid": "6074fa0b4ec4b9b0", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
| 302 |
+
{"build": "messytable_r2", "row_uid": "12616a97f218d54f", "split": "train", "reason": "prior_downsample", "step": "answer", "answer": "2"}
|
reports/s14_sample.tsv
ADDED
|
@@ -0,0 +1,201 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# row_uid task split -- the S14 spot-check rows; also flagged S14.sampled_qa
|
| 2 |
+
9cf58690a9f8408b multiview_counting train
|
| 3 |
+
9ba7d5ebb7e52811 multiview_counting train
|
| 4 |
+
a77755bc9db9ca06 multiview_counting train
|
| 5 |
+
026f91de9b853588 multiview_counting train
|
| 6 |
+
e8561872be0e90f9 multiview_counting train
|
| 7 |
+
ad53e044ba3a1b01 multiview_counting train
|
| 8 |
+
147b992609b57e4c multiview_counting train
|
| 9 |
+
d57e6caf22e4aa19 multiview_counting train
|
| 10 |
+
e1257ffc6671b108 multiview_counting train
|
| 11 |
+
384b7091ccf5736e multiview_counting train
|
| 12 |
+
957da47ef92882e5 multiview_counting train
|
| 13 |
+
d4e2dbbb86a867bd multiview_counting train
|
| 14 |
+
ef018a403fda8faa multiview_counting train
|
| 15 |
+
2d414680dce5a647 multiview_counting train
|
| 16 |
+
e7e0d1ff05980045 multiview_counting train
|
| 17 |
+
1d8c8f9d1a8947bd multiview_counting train
|
| 18 |
+
941adcc3aca45359 multiview_counting validation
|
| 19 |
+
8fd303f27a28fa64 multiview_counting train
|
| 20 |
+
c795df306aeafadc multiview_counting train
|
| 21 |
+
7eff5c476ae286ef multiview_counting train
|
| 22 |
+
20d7dacc5ee76605 multiview_counting train
|
| 23 |
+
a71ca7aee879116e multiview_counting train
|
| 24 |
+
9fb58814499fe75e multiview_counting train
|
| 25 |
+
0b06121ee7483a94 multiview_counting train
|
| 26 |
+
f15d07dade285f7f multiview_counting train
|
| 27 |
+
bf54a47ade84b455 multiview_counting train
|
| 28 |
+
b471022288ec548e multiview_counting train
|
| 29 |
+
f2f68afb185c1f82 multiview_counting train
|
| 30 |
+
dd5994571d767787 multiview_counting train
|
| 31 |
+
2b99da9805f5805d multiview_counting train
|
| 32 |
+
02a67ce0a2a0a5c2 multiview_counting train
|
| 33 |
+
59c6c6d13f981d51 multiview_counting train
|
| 34 |
+
2282ebb1adce16fa multiview_counting train
|
| 35 |
+
b1f2918db53bec1a multiview_counting train
|
| 36 |
+
cb33c2f530c34e2a multiview_counting train
|
| 37 |
+
6558e58a867b7b2f multiview_counting train
|
| 38 |
+
2cbc7d2bd04d6dfd multiview_counting train
|
| 39 |
+
89b8416a6ade06f2 multiview_counting train
|
| 40 |
+
4a0af87d04f9970b multiview_counting train
|
| 41 |
+
96a6f638839bd0ec multiview_counting train
|
| 42 |
+
fdf62ad4048145ee multiview_counting train
|
| 43 |
+
80e7ee0bee9efa4b multiview_counting train
|
| 44 |
+
98d32a32dd791d5d multiview_counting train
|
| 45 |
+
51eb62017ab57748 multiview_counting train
|
| 46 |
+
a69da82b1803d078 multiview_counting train
|
| 47 |
+
ccfe487d5f4c5b99 multiview_counting train
|
| 48 |
+
6c0fc67f7443ad45 multiview_counting train
|
| 49 |
+
29725ddba2bb62e5 multiview_counting train
|
| 50 |
+
8766237c24237f95 multiview_counting train
|
| 51 |
+
943c2f1331d799fe multiview_counting train
|
| 52 |
+
7d42366204c94469 multiview_counting train
|
| 53 |
+
f80d10b693342ae6 multiview_counting train
|
| 54 |
+
483dc4f74a0fbd99 multiview_counting train
|
| 55 |
+
79019354df8fbd21 multiview_counting train
|
| 56 |
+
0f60efa809a685cf multiview_counting train
|
| 57 |
+
1b45f56d810b208b multiview_counting train
|
| 58 |
+
e8ec80d669d48432 multiview_counting train
|
| 59 |
+
152f459c3f27a957 multiview_counting train
|
| 60 |
+
750cf88afa1a3d63 multiview_counting train
|
| 61 |
+
1147a2535488054b multiview_counting train
|
| 62 |
+
f0a1e596e18b23a5 multiview_counting train
|
| 63 |
+
1c0647b36af2fa1b multiview_counting train
|
| 64 |
+
36b1a0b3b239fc3d multiview_counting train
|
| 65 |
+
d451df59f9df1370 multiview_counting train
|
| 66 |
+
c0fb3000992c9064 multiview_counting train
|
| 67 |
+
019d03f17c6f739e multiview_counting train
|
| 68 |
+
90a1218a43601851 multiview_counting train
|
| 69 |
+
6ae89150a5977ca9 multiview_counting train
|
| 70 |
+
ff18b0bde8a9e7d6 multiview_counting train
|
| 71 |
+
fe8825168c0be432 multiview_counting train
|
| 72 |
+
098c77e03dd683af multiview_counting train
|
| 73 |
+
1b29db94dd9e8a1e multiview_counting train
|
| 74 |
+
c56d91ace337b9d5 multiview_counting train
|
| 75 |
+
052c6ed8f05a12ab multiview_counting train
|
| 76 |
+
a1a00f4574b559d7 multiview_counting train
|
| 77 |
+
afe31ad458b074cf multiview_counting train
|
| 78 |
+
791882c25f92ee09 multiview_counting train
|
| 79 |
+
754a2537e8ac6100 multiview_counting train
|
| 80 |
+
b2a4b92db9d26136 multiview_counting train
|
| 81 |
+
c7ebeb18e2b32d64 multiview_counting train
|
| 82 |
+
f90d638c082dde7a multiview_counting train
|
| 83 |
+
5cf9a3ed3e663387 multiview_counting train
|
| 84 |
+
ca4fa62b866823a3 multiview_counting train
|
| 85 |
+
6577e2847cf6fe08 multiview_counting train
|
| 86 |
+
3d9cf9dfdb2ab911 multiview_counting train
|
| 87 |
+
55f0cbfc762773c7 multiview_counting train
|
| 88 |
+
330171b08eb6b38f multiview_counting train
|
| 89 |
+
a77db4fdccdd4513 multiview_counting train
|
| 90 |
+
b1703558381ba571 multiview_counting train
|
| 91 |
+
87e95ad2e5350c7a multiview_counting train
|
| 92 |
+
4b9c1eaec279aa06 multiview_counting train
|
| 93 |
+
7d0380763fea913d multiview_counting train
|
| 94 |
+
ccf2d3d2d8500a6c multiview_counting train
|
| 95 |
+
bd7cd17148386382 multiview_counting train
|
| 96 |
+
dc7f5188781a82a5 multiview_counting train
|
| 97 |
+
8eab5a2344147122 multiview_counting train
|
| 98 |
+
3610b3b2e22547ee multiview_counting train
|
| 99 |
+
942fb989cd8a03ea multiview_counting train
|
| 100 |
+
db7e284ac0613413 multiview_counting train
|
| 101 |
+
1024539ee5b1640d multiview_counting train
|
| 102 |
+
a04c6037ef61e218 multiview_counting train
|
| 103 |
+
b6c10ccd432b6568 multiview_counting train
|
| 104 |
+
ea1d347c1d6a03e6 multiview_counting train
|
| 105 |
+
3eb5458570188ff5 multiview_counting train
|
| 106 |
+
36a63e9297b25a2e multiview_counting train
|
| 107 |
+
04a67bb3a5b049eb multiview_counting train
|
| 108 |
+
95f997f4d4f60f4c multiview_counting train
|
| 109 |
+
b8ed8dbfddf52e2a multiview_counting train
|
| 110 |
+
a2bc6c44bef6642f multiview_counting train
|
| 111 |
+
23e360cfd2a67117 multiview_counting train
|
| 112 |
+
126ea2d6c09fa49e multiview_counting train
|
| 113 |
+
1e83ae1114678809 multiview_counting train
|
| 114 |
+
b3b74e6cb46e441a multiview_counting train
|
| 115 |
+
4e696536f599aa98 multiview_counting train
|
| 116 |
+
11f56bb6aa824f49 multiview_counting train
|
| 117 |
+
7a7ec6670fffd8c4 multiview_counting train
|
| 118 |
+
29c2258390a5b8e4 multiview_counting train
|
| 119 |
+
7b73facc8da92d72 multiview_counting train
|
| 120 |
+
332c50270e923e99 multiview_counting train
|
| 121 |
+
f80df91769d3a327 multiview_counting train
|
| 122 |
+
ace1360d50aa9d59 multiview_counting train
|
| 123 |
+
e4520bd4a358edd9 multiview_counting train
|
| 124 |
+
068b99f017116d49 multiview_counting train
|
| 125 |
+
ce791f59c9f7c46d multiview_counting train
|
| 126 |
+
66c75578ca40aeda multiview_counting train
|
| 127 |
+
0c165d166d5026b9 multiview_counting train
|
| 128 |
+
c35ff36846d9dfa3 multiview_counting train
|
| 129 |
+
d8e68eb8e430f448 multiview_counting train
|
| 130 |
+
dcf803483dc6b720 multiview_counting train
|
| 131 |
+
c51817333fd07ad6 multiview_counting train
|
| 132 |
+
e5b47d01b3cd8ac9 multiview_counting train
|
| 133 |
+
c5792cf975638440 multiview_counting train
|
| 134 |
+
928d424adb814f94 multiview_counting train
|
| 135 |
+
ebdea8c795e1de93 multiview_counting train
|
| 136 |
+
9e518d8a4084938f multiview_counting train
|
| 137 |
+
fc629b777acbd8d0 multiview_counting train
|
| 138 |
+
a7edf4c2ad1cd0ab multiview_counting train
|
| 139 |
+
e03d229c79bfe625 multiview_counting train
|
| 140 |
+
a17481aab80822a3 multiview_counting train
|
| 141 |
+
5fa65e4fc9e1136f multiview_counting train
|
| 142 |
+
b9133916acaf7c42 multiview_counting train
|
| 143 |
+
5b37af30217261f9 multiview_counting train
|
| 144 |
+
5db4a10893cbfc4d multiview_counting train
|
| 145 |
+
2446081b66203ea2 multiview_counting train
|
| 146 |
+
3964e66a71c6ed03 multiview_counting train
|
| 147 |
+
e40c10910b877c6f multiview_counting train
|
| 148 |
+
01491f33ea9a2a0e multiview_counting train
|
| 149 |
+
253121d19327e2b6 multiview_counting train
|
| 150 |
+
02ce91e631f767fa multiview_counting train
|
| 151 |
+
4f983a8b13a5f19b multiview_counting train
|
| 152 |
+
f868bcccf4badb5f multiview_counting train
|
| 153 |
+
48da8bf6394387d3 multiview_counting train
|
| 154 |
+
9d7f517258e6965f multiview_counting train
|
| 155 |
+
5a62dd6bad894e4c multiview_counting train
|
| 156 |
+
4f75aba7d17328ab multiview_counting train
|
| 157 |
+
abfea2135946075d multiview_counting train
|
| 158 |
+
2ccc38b3416c3eb1 multiview_counting train
|
| 159 |
+
20d34907f0ff8871 multiview_counting train
|
| 160 |
+
d3c61f7f5c78df9c multiview_counting train
|
| 161 |
+
ab39035db0511238 multiview_counting train
|
| 162 |
+
7f871441f666358d multiview_counting train
|
| 163 |
+
5feb6ac3f378c028 multiview_counting train
|
| 164 |
+
0545e7a4b3a1b1ed multiview_counting train
|
| 165 |
+
600a333a160c7d71 multiview_counting train
|
| 166 |
+
1191e2f6140e28b4 multiview_counting train
|
| 167 |
+
dc94284cf5d283fc multiview_counting train
|
| 168 |
+
78669336aa9e5ccc multiview_counting train
|
| 169 |
+
3aae02079c6bb1bd multiview_counting train
|
| 170 |
+
a9dfbdfcdc765163 multiview_counting train
|
| 171 |
+
4dc98bbbe9aa1830 multiview_counting train
|
| 172 |
+
5ff940d374be9404 multiview_counting train
|
| 173 |
+
9a80cb6f0cdcb789 multiview_counting validation
|
| 174 |
+
6b4801d25fb6705c multiview_counting train
|
| 175 |
+
2d308f1341ba5d96 multiview_counting validation
|
| 176 |
+
4a24cbe380f1317e multiview_counting train
|
| 177 |
+
cb24d10fca692705 multiview_counting train
|
| 178 |
+
987d20b6dabd0c12 multiview_counting train
|
| 179 |
+
e86ed7c69d24e5db multiview_counting train
|
| 180 |
+
d6669c792be96165 multiview_counting train
|
| 181 |
+
7b1c3248ec4b0447 multiview_counting train
|
| 182 |
+
11801f793974a8c6 multiview_counting train
|
| 183 |
+
30349adc8b7689ed multiview_counting train
|
| 184 |
+
9c6a34ef2b338470 multiview_counting train
|
| 185 |
+
e5ac9cf7d8aa4e29 multiview_counting train
|
| 186 |
+
e3c186f8251bd595 multiview_counting train
|
| 187 |
+
f5ee9442c22ece39 multiview_counting train
|
| 188 |
+
f637dd2d3bcb129f multiview_counting train
|
| 189 |
+
81ad54cb7c1ff047 multiview_counting train
|
| 190 |
+
007e59eab2b51724 multiview_counting train
|
| 191 |
+
7c4f43772f8bdcc7 multiview_counting train
|
| 192 |
+
a5a31815680f66b4 multiview_counting train
|
| 193 |
+
c29c589cb46d5629 multiview_counting train
|
| 194 |
+
63d563267de64a38 multiview_counting train
|
| 195 |
+
165963a0cc23d5fc multiview_counting train
|
| 196 |
+
5aff2fd723325207 multiview_counting train
|
| 197 |
+
7c74c9637fb65d82 multiview_counting train
|
| 198 |
+
797cac9380330349 multiview_counting train
|
| 199 |
+
2f702c41f7714de3 multiview_counting train
|
| 200 |
+
796f18b08ac01c9a multiview_counting train
|
| 201 |
+
fe2011d462bdc5b5 multiview_counting train
|
train/chunk_00000.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:f387311a7f86e8650167ddbc2179cb90017d1ad2b15c67763f70760a19452bbe
|
| 3 |
+
size 439073826
|
validation/chunk_00000.parquet
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:b472e51ce2026e3966c192f124bc0f008031049c1db89e323ae5ffadf4826cab
|
| 3 |
+
size 8404021
|