Papers › Cross-Care: Assessing the Healthcare Implications of Pre-training Data on Language Model Bias

Cross-Care: Assessing the Healthcare Implications of Pre-training Data on Language Model Bias

9 May 2024arXiv:2405.05506archive 2025-07-28

Shan Chen, Jack Gallifant, Mingye Gao, Pedro Moreira, Nikolaj Munch, Ajay Muthukkumar, Arvind Rajan, Jaya Kolluri, Amelia Fiske, Janna Hastings, Hugo Aerts, Brian Anthony, Leo Anthony Celi, William G. La Cava, Danielle S. Bitterman

Large language models (LLMs) are increasingly essential in processing natural languages, yet their application is frequently compromised by biases and inaccuracies originating in their training data. In this study, we introduce Cross-Care, the first benchmark framework dedicated to assessing biases and real world knowledge in LLMs, specifically focusing on the representation of disease prevalence across diverse demographic groups. We systematically evaluate how demographic biases embedded in pre-training corpora like ThePile influence the outputs of LLMs. We expose and quantify discrepancies by juxtaposing these biases against actual disease prevalences in various U.S. demographic groups. Our results highlight substantial misalignment between LLM representation of disease prevalence and real disease prevalence rates across demographic subgroups, indicating a pronounced risk of bias propagation and a lack of real-world grounding for medical applications of LLMs. Furthermore, we observe that various alignment methods minimally resolve inconsistencies in the models' representation of disease prevalence across different languages. For further exploration and analysis, we make all data and a data visualization tool available at: www.crosscare.net.

PaperPDFCodeCode Syntology ran

In Syntology View this paper on Syntology, its page in Syntology's graph. That page adds which of its functions are also present in other papers' code Syntology holds, with dates, and its dated run record as JSON and a README badge. It also lists the repositories linked to the paper, the abstract and the calls for agents.

Open this paper in Syntology's Atlas, the map of the papers in Syntology's graph and their citations.

For agents, on Syntology's MCP service (how to connect):

Code

Syntology 20 code samples harvested from 1 repository linked to this paper: 16 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 6 where Syntology's instrument failed) · 4 unverified. Where Syntology's instrument failed: 6 where our contract draft was wrong; failures of Syntology's instrument, not of the code.

By repository: official repository: 20 samples from 1 repository, 16 ran (10 with no instrument failure; 6 where Syntology's instrument failed). The run record, sample by sample. “Ran” means executed on a synthesized input, not that the code is correct or reproduces the paper.

Levels written on 2026-09-24 (UTC): the dates Syntology's graph wrote each sample's verification level, not when the code executed.

shan23chen/cross-care officialmentioned in papermentioned on GitHubtfMITSyntology: 20 samples, 16 ran (6 instrument-failed) · 4 unverified report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

20 samples harvested: 16 ran (of which 0 constructed an object rather than computing a result; 10 with no instrument failure: 1 honoured, 0 violated, 9 with no contract checked; 6 where Syntology's instrument failed) · 4 unverified. “Where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. That date is not when the samples ran.

1ran · honoured contract
6ran · our draft was wrong
9ran
4unverified

Licence: 0 of the 20 samples are pointer only, meaning Syntology does not serve that copy's text. This page shows no code text for any sample; each one links to its file in the repository.

Harvested from shan23chen/cross-care. “Ran” means the sample executed on a synthesized input. It does not mean the output is correct, and nothing here reproduces the paper's results. A line beginning “fixture:” under a sample that ran gives the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right. “Level recorded” gives the UTC date when Syntology's graph wrote the sample's verification level, not when the code executed; where the level was written before Syntology's graph kept that date, the row says “level recorded, date not recorded”. “Honoured” and “violated” refer to a contract Syntology drafted from the code itself. “Our draft was wrong” and “the contract check could not run” are failures of Syntology's instrument, not of the code: the sample ran on Syntology's synthesized input, and then the drafted contract did not describe it, or the check itself raised an error (RAISES), lacked a dependency (DEP_MISSING) or ran out of time (TIMEOUT); hover the label for which. A metamorphic tier, where shown, is evidence of the output's shape, not that the sample implements the method.

Each sample ends with its code_sha256. A sample's code_sha256 is Syntology's identity for that exact code. It is printed in full (16 hexadecimal characters for every sample in this build), and get_code(code_sha256="…") takes it exactly as printed. An agent calls it on Syntology's MCP service (how to connect); click an identity to copy that call.

Repository labels, per sample. official repository: The archive marks this repository official for the paper. named in the paper: The archive records that the paper mentions this repository; it is not marked official. community (archive-listed): In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper. found in paper text by Syntology: Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted. community: Not in the archive's code links for this paper; a community repository Syntology harvested. Samples are grouped by repository: a repository marked official first, then by how directly the repository is tied to the paper, then in the archive's order; never by how many ran. Licence labels name the repository's licence as recorded at harvest. “Pointer only” means Syntology does not serve that copy's text, for one of six reasons: no licence file was found; the licence was not identified; the licence is recorded as permissive but that copy's record is not marked cleared; the licence is outside the permissive list Syntology serves text under (MIT, Apache-2.0, BSD and similar); this paper's copy was not located, so its licence is not recorded; or this paper's copy is cleared, but the stored copy of the same code that Syntology serves is not. Some licences outside that list permit redistribution, such as WTFPL, and GPL-3.0 under its conditions; they are simply not on the list. Hover a licence label for the reason. File links open the file on GitHub at the default branch, which may have changed since the harvest.

A function or class name links to its page when the name occurs in the harvested code of at least 5 papers; the name is a string, not an identity.

shan23chen/cross-care official repository 20 samples, 16 ran (6 instrument-failed) · 4 unverified
calculate_disease_by_group co_occurrence_generate/co_occurrence_analysis.py ran MIT (permissive) · 4d08d3d7dced2ae0 · report fixture: signature stack_dataframe:DataFrame/0/?, medical_dict:dict/0/?, gender_dict:dict/0/?, racial_dict:dict/0/?, drug_dict:dict/0/? · returned tuple · level recorded 2026-09-24
calculate_log_softmax_batch_non_seq2seq logits_generate/hf_model.py ran MIT (permissive) · 9a58c6dbe3abe198 · report fixture: signature texts:list/1/str, tokenizer:callable/0/?, model:callable/0/?, device:str/0/str · returned list · level recorded 2026-09-24
co_occurrence_within_window co_occurrence_generate/co_occurrence_analysis.py ran MIT (permissive) · 8bc7cb8d8bef7fc5 · report fixture: signature data:list/1/str, medical_dict:dict/0/?, racial_dict:dict/0/?, gender_dict:dict/0/?, drug_dict:dict/0/?, window_size:int/0/int, aggregate:bool/0/bool · returned tuple · level recorded 2026-09-24
create_keyword_pattern co_occurrence_generate/co_occ_analysis.py ran · our draft was wrong MIT (permissive) · 6f934cc1b439e9f8 · report fixture: signature keywords:list/1/str · returned Pattern · level recorded 2026-09-24
create_keyword_pattern co_occurrence_generate/co_occurrence_analysis.py ran · our draft was wrong MIT (permissive) · dd0d546bb4bf0215 · report fixture: signature keywords:list/1/str · returned Pattern · level recorded 2026-09-24
create_keyword_pattern co_occurrence_generate/jsonl_data_filtering_c4.py ran · our draft was wrong MIT (permissive) · 88ce89805a8ffe04 · report fixture: signature keywords:list/1/str · returned Pattern · level recorded 2026-09-24
extract_total_count_from_message co_occurrence_generate/co_occurrence_infinigram_document_counts.py ran · honoured contract fingerprinted MIT (permissive) · 59281530c199bbdd · report fixture: signature message:str/0/str · returned int · level recorded 2026-09-24
find_co_occurrences co_occurrence_generate/co_occ_analysis.py ran MIT (permissive) · 1194ae31aea58742 · report fixture: signature text:str/0/str, medical_patterns:dict/0/?, racial_patterns:dict/0/?, gender_patterns:dict/0/?, drug_patterns:dict/0/?, window_size:int/0/int · returned dict · level recorded 2026-09-24
find_duplicates_in_file co_occurrence_generate/check_keywords.py ran MIT (permissive) · b894e1ed19d48aea · report fixture: signature file_path:str/0/str, list_name:str/0/str · returned list · level recorded 2026-09-24
generate_case_variations co_occurrence_generate/co_occurrence_infinigram.py ran · our draft was wrong MIT (permissive) · f1d7a3870630867d · report fixture: signature term:str/0/str · returned set · level recorded 2026-09-24
process_file in 33 papers co_occurrence_generate/jsonl_data_filtering.py ran MIT (permissive) · 293bc1ef76e375ef · report fixture: signature file_path:str/0/str, medical_patterns:dict/0/?, racial_patterns:dict/0/?, gender_patterns:dict/0/?, drug_patterns:dict/0/?, cancer_patterns:dict/0/?, metadata_keys:list/1/str, remove_latex:bool/0/bool · returned tuple · level recorded 2026-09-24
process_file in 33 papers co_occurrence_generate/jsonl_data_filtering_c4.py ran MIT (permissive) · fe4d91f2f027fe4e · report fixture: returned tuple · level recorded 2026-09-24
remove_latex_commands co_occurrence_generate/jsonl_data_filtering.py ran · our draft was wrong fingerprinted MIT (permissive) · c4a0e7d1de5f3a1c · report fixture: signature s:str/0/str · returned str · level recorded 2026-09-24
remove_latex_commands co_occurrence_generate/jsonl_data_filtering_c4.py ran · our draft was wrong fingerprinted MIT (permissive) · a56f744d089c3d45 · report fixture: signature s:str/0/str · returned str · level recorded 2026-09-24
search_documents_count co_occurrence_generate/disease_document_count_infinigram.py ran MIT (permissive) · 94cf61296c1ba3b8 · report fixture: signature term:str/0/str, corpora:list/1/str, maxnum:int/0/int · returned dict · level recorded 2026-09-24
tokenize_for_seq2seq logits_generate/hf_model.py ran MIT (permissive) · f855d60484b63f8c · report fixture: signature texts:list/1/str, model:callable/0/?, tokenizer:callable/0/?, device:str/0/str, target_text:str/0/str · returned tuple · level recorded 2026-09-24
calculate_log_softmax_batch_seq2seq logits_generate/hf_model.py unverified MIT (permissive) · 23ba4360a5bd3294 · report
count_term co_occurrence_generate/co_occurrence_infinigram.py unverified MIT (permissive) · 4ea6258eabae7b12 · report
process_texts_in_parallel co_occurrence_generate/co_occ_analysis.py unverified MIT (permissive) · 8d57c440e1ad8631 · report
search_documents_for_corpus co_occurrence_generate/co_occurrence_infinigram_document_counts.py unverified MIT (permissive) · bf0bc2de96cec2ca · report

Tasks

Data VisualizationLanguage ModelingLanguage ModellingWorld Knowledge

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections