Rebuild on the benchmark's own answer protocol; columns id, query, response, scoring_context first

README.md

@@ -30,7 +30,7 @@ configs:
 
 # MuMOInstruct, hub-native
 
-`NingLab/MuMOInstruct` (`f3ca492caaca6ecb33b1fcff65d882f10acc3297`), restricted to the paper's own IND/OOD evaluation split (the ten declared property combinations) and restructured into `ind`/`ood` subsets, with the unified columns `id`, `query`, `ground_truth`, `response`, `subtask` and `scoring_context` added beside every original one. `ground_truth` and `response` are blank for every row; the oracle scores a generated beam against property predictors, not a stored answer. See `scripts/datasets/README.md` in the AutoDataSci repository for what those columns mean.
+`NingLab/MuMOInstruct` (`f3ca492caaca6ecb33b1fcff65d882f10acc3297`), restricted to the paper's own IND/OOD evaluation split (the ten declared property combinations) and restructured into `ind`/`ood` subsets, with the unified columns `id`, `query`, `ground_truth`, `response`, `subtask` and `scoring_context` added beside every original one. `ground_truth` is the train split's own `target_smiles` and `response` its tagged form; both are blank for every test row, which upstream carries no target for -- the oracle scores the one candidate's property change against the source molecule, not a stored answer. See `scripts/datasets/README.md` in the AutoDataSci repository for what those columns mean.
 
 Row counts: {"train": 19841, "test": 7810}.
 

all/test.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:dffa85b13a2ecb971677cca07c96a09e5c55c0e54acf635bba795d7264c52e2a
-size 1340377
+oid sha256:e222b6a30ecb5082f62f98f607acbd07b7bf2910e4d8e696e7ddd5ee0e01a1e6
+size 1339383

all/train.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:236ad0b14562c06a5e435c8fa3ed967a339b4d32eb2a363b53931b2c2c5de7e1
-size 4364434
+oid sha256:2fb66a722ab2bff69e78fa07158c846400b897ee3bd36562551dcf1a259e3abc
+size 4829755

ind/test.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:171252ee0701cb294a0a4f268bf515acdae0e00f6a1078141b4972925917a131
-size 881904
+oid sha256:597565ef06d332f0ea314d3ec1da0506502fdc435a9963ba5a6f25cb6e91119e
+size 881149

ind/train.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:4215b4982eae17b5d981399a57caae190c8ad11c4019d2b0d896362d314769f7
-size 3306684
+oid sha256:572698ae3dc91c9a03cffabbcca38f75b68d1fc1452eb0189b7979901902d2ec
+size 3662963

ood/test.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:d1d10b7869d7c61fb84d06d6a727349037bba84710767f6c3f121724e31a5126
-size 494559
+oid sha256:b6cc5265d72c31634682cb0e3faaec10c77565cd7a8744d6f90444941481ae06
+size 494192

ood/train.parquet

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8128489e4db41e2e2af7afc011f599efd4429e6360a0fa47d0fcca14e134f8dd
-size 1081062
+oid sha256:63f7f83ffdbe8a35fe8961592f25fda7237ea2d2dba03371adff29afbc6721a9
+size 1238296