{"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/query-model","entry":"query_model","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":7,"n_papers_ran":1,"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":7,"n_samples_ran":1,"n_samples_fingerprinted":0,"n_places":7,"n_places_pointer_only":2,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":1,"ran_fixture":0,"ran":0,"unverified":6},"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":"2503.15092","paper":"/paper/towards-understanding-the-safety-boundaries","title":"Towards Understanding the Safety Boundaries of DeepSeek Models: Evaluation and Findings","date":"2025-03-19","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"ny1024/deepseek-safety-eval","path":"eval/base/deepseek_r1.py","file_url":"https://github.com/ny1024/deepseek-safety-eval/blob/HEAD/eval/base/deepseek_r1.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"30d1ee7bb73c4e61","mcp_get_code":{"code_sha256":"30d1ee7bb73c4e61"}},{"arxiv_id":"2501.04227","paper":"/paper/agent-laboratory-using-llm-agents-as-research","title":"Agent Laboratory: Using LLM Agents as Research Assistants","date":"2025-01-08","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"Masao-Taketani/LocalAgentLaboratory","path":"inference.py","file_url":"https://github.com/Masao-Taketani/LocalAgentLaboratory/blob/HEAD/inference.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"528c51546feae850","mcp_get_code":{"code_sha256":"528c51546feae850"}},{"arxiv_id":"2411.00056","paper":null,"title":"arXiv:2411.00056","date":null,"month_inferred_from_arxiv_id":"2024-11","title_source":null,"repo":"DarianRodriguez/NegVerse","path":"negator/Evaluation/evaluate.py","file_url":"https://github.com/DarianRodriguez/NegVerse/blob/HEAD/negator/Evaluation/evaluate.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"ecdd4dce2f7ddb5c","mcp_get_code":{"code_sha256":"ecdd4dce2f7ddb5c"}},{"arxiv_id":"2406.06608","paper":"/paper/the-prompt-report-a-systematic-survey-of","title":"The Prompt Report: A Systematic Survey of Prompting Techniques","date":"2024-06-06","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"trigaten/Prompt_Systematic_Review","path":"src/prompt_systematic_review/experiments/benchmarking.py","file_url":"https://github.com/trigaten/Prompt_Systematic_Review/blob/HEAD/src/prompt_systematic_review/experiments/benchmarking.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"590036abca2df4b6","mcp_get_code":{"code_sha256":"590036abca2df4b6"}},{"arxiv_id":"2310.17924","paper":"/paper/soul-towards-sentiment-and-opinion","title":"SOUL: Towards Sentiment and Opinion Understanding of Language","date":"2023-10-27","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"DAMO-NLP-SG/SOUL","path":"code/llm.py","file_url":"https://github.com/DAMO-NLP-SG/SOUL/blob/HEAD/code/llm.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"182e039b297c5f9b","mcp_get_code":{"code_sha256":"182e039b297c5f9b"}},{"arxiv_id":"2025.naacl-long.65","paper":null,"title":"arXiv:2025.naacl-long.65","date":null,"month_inferred_from_arxiv_id":null,"title_source":null,"repo":"Zehong-Wang/TANS","path":"TANS/preprocess/query_llms.py","file_url":"https://github.com/Zehong-Wang/TANS/blob/HEAD/TANS/preprocess/query_llms.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"eb4ab71f22499989","mcp_get_code":{"code_sha256":"eb4ab71f22499989"}},{"arxiv_id":"2024.findings-emnlp.407","paper":null,"title":"arXiv:2024.findings-emnlp.407","date":null,"month_inferred_from_arxiv_id":null,"title_source":null,"repo":"amayuelas/multi-agent-attack","path":"multiagent_debate/commons.py","file_url":"https://github.com/amayuelas/multi-agent-attack/blob/HEAD/multiagent_debate/commons.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"1730bf85e8c595b0","mcp_get_code":{"code_sha256":"1730bf85e8c595b0"}}]}