DPOT: Auto-Regressive Denoising Operator Transformer for Large-Scale PDE Pre-Training
Paper • 2403.03542 • Published
arxiv_doi stringlengths 24 31 | arxiv_id stringlengths 9 16 | citation_count int64 1 44.9k | cited_by listlengths 1 44.9k | reference_count int64 1 45.1k |
|---|---|---|---|---|
10.48550/arXiv.1412.6980 | 1412.6980 | 44,907 | [{"doi":"10.1001/jamadermatol.2022.5415","matches":[{"provenance":"mined","raw_match":"arxiv.org/abs(...TRUNCATED) | 45,061 |
10.48550/arXiv.1706.03762 | 1706.03762 | 36,818 | [{"doi":"10.1001/jama.2023.25057","matches":[{"provenance":"mined","raw_match":"arxiv.org/abs/1706.0(...TRUNCATED) | 37,014 |
10.48550/arXiv.1409.1556 | 1409.1556 | 31,493 | [{"doi":"10.1001/jamanetworkopen.2018.8029","matches":[{"provenance":"mined","raw_match":"arxiv.org/(...TRUNCATED) | 31,709 |
10.48550/arXiv.1810.04805 | 1810.04805 | 23,043 | [{"doi":"10.1001/jamacardio.2023.3142","matches":[{"provenance":"mined","raw_match":"arxiv.org/abs/1(...TRUNCATED) | 23,232 |
10.48550/arXiv.2010.11929 | 2010.11929 | 21,677 | [{"doi":"10.1001/jamanetworkopen.2025.13921","matches":[{"provenance":"mined","raw_match":"10.48550/(...TRUNCATED) | 21,845 |
10.48550/arXiv.1804.02767 | 1804.02767 | 10,978 | [{"doi":"10.1001/jamanetworkopen.2022.9960","matches":[{"provenance":"mined","raw_match":"arxiv.org/(...TRUNCATED) | 11,047 |
10.48550/arXiv.1704.04861 | 1704.04861 | 10,406 | [{"doi":"10.1001/jamanetworkopen.2019.0442","matches":[{"provenance":"mined","raw_match":"arxiv.org/(...TRUNCATED) | 10,472 |
10.48550/arXiv.1609.02907 | 1609.02907 | 9,983 | [{"doi":"10.1002/9781119390060.ch9","matches":[{"provenance":"mined","raw_match":"arXiv:1609.02907",(...TRUNCATED) | 10,044 |
10.48550/arXiv.1301.3781 | 1301.3781 | 9,546 | [{"doi":"10.1001/jamanetworkopen.2025.6359","matches":[{"provenance":"mined","raw_match":"10.48550/a(...TRUNCATED) | 9,612 |
10.48550/arXiv.1907.11692 | 1907.11692 | 9,387 | [{"doi":"10.1001/jamanetworkopen.2023.3079","matches":[{"provenance":"mined","raw_match":"10.48550/a(...TRUNCATED) | 9,453 |
End of preview. Expand in Data Studio
Crossref arXiv Citations
A dataset of arXiv preprints and their citations extracted from Crossref metadata, validated against DataCite records.
Dataset Description
This dataset maps arXiv works to the works in Crossref that cite them. Each record represents an arXiv preprint with all known citations from Crossref-registered works.
Built from the Crossref Metadata Plus monthly snapshot 2026-07 and the
DataCite monthly data file 2026-07 with
crossref-citation-extraction
version 2.2.0.
Dataset Configurations
| Config | Description | arXiv Works | Citing Works | Reference Matches |
|---|---|---|---|---|
all |
All validated citations (default) | 1,042,011 | 5,999,691 | 6,028,756 |
asserted |
Citations where the DOI was explicitly provided by the publisher or matched by Crossref | 160,091 | 452,505 | 454,190 |
mined |
Citations mined from reference metadata without a publisher or Crossref DOI assertion | 998,112 | 5,548,053 | 5,574,566 |
Loading the Dataset
from datasets import load_dataset
dataset = load_dataset("cometadata/crossref-arxiv-citations")
asserted = load_dataset("cometadata/crossref-arxiv-citations", "asserted")
mined = load_dataset("cometadata/crossref-arxiv-citations", "mined")
Data Schema
Each record contains:
| Field | Type | Description |
|---|---|---|
arxiv_doi |
string | The DOI for the arXiv work (format: 10.48550/arXiv.{id}) |
arxiv_id |
string | The arXiv identifier (e.g., 1412.6980) |
reference_count |
integer | Total number of reference match instances |
citation_count |
integer | Number of unique citing works |
cited_by |
array | List of citing work objects |
Each citing work object in cited_by contains:
| Field | Type | Description |
|---|---|---|
doi |
string | DOI of the citing work |
provenance |
string | Highest-confidence provenance among this work's matches (publisher > crossref > mined) |
matches |
array | Reference matches from this citing work |
Each match contains:
| Field | Type | Description |
|---|---|---|
provenance |
string | How this match was obtained (see below) |
raw_match |
string | The exact reference text that identified the arXiv work (a substring, or a whole structured value): an arXiv identifier, a DOI, or a URL such as the work's registered landing page |
reference |
struct | The Crossref reference metadata (string-valued fields; absent keys are null) |
Provenance Values
publisher- the reference's structuredDOIfield holds the arXiv DOI withdoi-asserted-by: publishercrossref- the reference's structuredDOIfield holds the arXiv DOI withdoi-asserted-by: crossrefmined- any other match: an arXiv identifier, DOI, or URL found in the reference'sDOI,URL,article-title,journal-title, orunstructuredfield (a structuredDOIcounts as mined unlessdoi-asserted-byispublisherorcrossref)
The asserted config contains only publisher and crossref matches; the
mined config contains only mined matches. A citing work appears in a
config only with its matches of that config's provenances, and all counts are
recomputed per config.
Extraction Process
arXiv references are extracted from the Crossref snapshot by scanning
reference metadata for arXiv identifiers in modern (arXiv:2403.03542),
legacy (arXiv:cs.DM/9910013), DOI (10.48550/arXiv.2403.03542), and URL
(arxiv.org/abs/2403.03542) forms. Identifiers are normalized (lowercase,
version suffixes removed). Extracted arXiv DOIs are validated against an
index of DOIs built from the DataCite monthly data file 2026-07;
only arXiv works registered in DataCite are included. When the dataset is
built by the full DataCite pipeline, an arXiv work can also be matched by its
DOI anywhere in the reference or by its registered DataCite landing-page URL.
Data Sources
- Crossref Metadata Plus monthly snapshot
2026-07 - DataCite monthly data file
2026-07 - Extraction tools: cometadata/crossref-arxiv-citation-extraction
Limitations
- Only includes citations registered in Crossref metadata
- Reference parsing depends on successful identifier extraction from reference metadata
- The
minedsubset may contain lower-quality matches from noisy reference text
License
This dataset is released under CC0 1.0 Universal.
Citation
@dataset{crossref_arxiv_citations,
title = {Crossref arXiv Citations},
author = {Cometadata},
publisher = {Hugging Face},
url = {https://huggingface.co/datasets/cometadata/crossref-arxiv-citations}
}
- Downloads last month
- 75