{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/code/load-system-prompt","entry":"load_system_prompt","source":"Syntology graph, per-sample; not an archive number","read_at":"2026-09-24T18:15:14+00:00","claim":"Names are grouped by exact entry-name string. Same-named routines are NOT asserted to be equivalent; 'ran' means executed on a synthesized fixture, not correctness. n_samples_ran = sum of by_status over every status except 'unverified' (ran_draft_wrong and ran_fixture are failures of Syntology's instrument, not of the code); n_papers_ran = papers with at least one such sample.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"},"n_papers":6,"n_papers_ran":2,"units":"n_samples, n_samples_ran, n_samples_fingerprinted and by_status count distinct code bodies (code_sha256); n_places and n_places_pointer_only count places, one per (paper, code body) pair, which is also the unit of the samples list","n_samples":6,"n_samples_ran":2,"n_samples_fingerprinted":1,"n_places":6,"n_places_pointer_only":3,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":1,"ran_fixture":0,"ran":1,"unverified":4},"syntology":{"atlas_url":null,"mcp":null,"mcp_per_sample":{"tool":"get_code","arguments_in":"samples[].mcp_get_code"},"developers":"https://syntology.ai/developers"},"samples":[{"arxiv_id":"2608.17809","paper":"/paper/arxiv-2608-17809","title":"Whether LLMs Can Navigate Beliefs and Facts Depends on How You Phrase It","date":null,"month_inferred_from_arxiv_id":"2026-08","title_source":"syntology","repo":"ngqm/belief-fact-phrasing","path":"src/eval/run_verb_eval_batched.py","file_url":"https://github.com/ngqm/belief-fact-phrasing/blob/HEAD/src/eval/run_verb_eval_batched.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"5f8cfc94998b1389","mcp_get_code":{"code_sha256":"5f8cfc94998b1389"}},{"arxiv_id":"2605.13412","paper":"/paper/arxiv-2605-13412","title":"LLMs as annotators of credibility assessment in Danish asylum decisions: evaluating classification performance and errors beyond aggregated metrics","date":null,"month_inferred_from_arxiv_id":"2026-05","title_source":"syntology","repo":"glhr/RAB-Cred","path":"llm_annotation/llm_utils.py","file_url":"https://github.com/glhr/RAB-Cred/blob/HEAD/llm_annotation/llm_utils.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"b449ec0c812f86f8","mcp_get_code":{"code_sha256":"b449ec0c812f86f8"}},{"arxiv_id":"2507.09580","paper":null,"title":"arXiv:2507.09580","date":null,"month_inferred_from_arxiv_id":"2025-07","title_source":null,"repo":"wangyu-ovo/aicrypto-agent","path":"run_choice_question.py","file_url":"https://github.com/wangyu-ovo/aicrypto-agent/blob/HEAD/run_choice_question.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"515e2108a26ccb36","mcp_get_code":{"code_sha256":"515e2108a26ccb36"}},{"arxiv_id":"2502.02673","paper":"/paper/medrax-medical-reasoning-agent-for-chest-x","title":"MedRAX: Medical Reasoning Agent for Chest X-ray","date":"2025-02-04","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"bowang-lab/medrax","path":"medrax/utils/utils.py","file_url":"https://github.com/bowang-lab/medrax/blob/HEAD/medrax/utils/utils.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"a76ec818e2e53ab4","mcp_get_code":{"code_sha256":"a76ec818e2e53ab4"}},{"arxiv_id":"2501.09409","paper":"/paper/mgente-a-multilingual-resource-for-gender","title":"mGeNTE: A Multilingual Resource for Gender-Neutral Language and Translation","date":"2025-01-16","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"g8a9/mgente-gap","path":"src/gnt_eval/local_evaluate.py","file_url":"https://github.com/g8a9/mgente-gap/blob/HEAD/src/gnt_eval/local_evaluate.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"9bfc3913b678e212","mcp_get_code":{"code_sha256":"9bfc3913b678e212"}},{"arxiv_id":"2311.08718","paper":"/paper/decomposing-uncertainty-for-large-language","title":"Decomposing Uncertainty for Large Language Models through Input Clarification Ensembling","date":"2023-11-15","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"ucsb-nlp-chang/llm_uncertainty","path":"forward.py","file_url":"https://github.com/ucsb-nlp-chang/llm_uncertainty/blob/HEAD/forward.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"62eddadce872572b","mcp_get_code":{"code_sha256":"62eddadce872572b"}}]}