{"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/process-csv","entry":"process_csv","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":5,"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":5,"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":"2605.04208","paper":"/paper/arxiv-2605-04208","title":"Nsanku: Evaluating Zero-Shot Translation Performance of LLMs for Ghanaian Languages","date":null,"month_inferred_from_arxiv_id":"2026-05","title_source":"syntology","repo":"GhanaNLP/nsanku","path":"repo/main-with-similarity-bk.py","file_url":"https://github.com/GhanaNLP/nsanku/blob/HEAD/repo/main-with-similarity-bk.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"dceffe868db232be","mcp_get_code":{"code_sha256":"dceffe868db232be"}},{"arxiv_id":"2601.16644","paper":"/paper/arxiv-2601-16644","title":"Sycophancy Hides Linearly in the Attention Heads","date":null,"month_inferred_from_arxiv_id":"2026-01","title_source":"syntology","repo":"rifoagenadi/sycophancy","path":"extension/judge.py","file_url":"https://github.com/rifoagenadi/sycophancy/blob/HEAD/extension/judge.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"97c9e69ad485811e","mcp_get_code":{"code_sha256":"97c9e69ad485811e"}},{"arxiv_id":"2408.11053","paper":"/paper/revisiting-verilogeval-newer-llms-in-context","title":"Revisiting VerilogEval: A Year of Improvements in Large-Language Models for Hardware Code Generation","date":"2024-08-20","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"nvlabs/verilog-eval","path":"count_failures.py","file_url":"https://github.com/nvlabs/verilog-eval/blob/HEAD/count_failures.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NOASSERTION","inline_ok":false,"code_sha256_prefix":"0b9143f06db3374a","mcp_get_code":{"code_sha256":"0b9143f06db3374a"}},{"arxiv_id":"2406.15885","paper":"/paper/the-music-maestro-or-the-musically-challenged","title":"The Music Maestro or The Musically Challenged, A Massive Music Evaluation Benchmark for Large Language Models","date":"2024-06-22","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"zcli-charlie/ziqi-eval","path":"src/five-shot.py","file_url":"https://github.com/zcli-charlie/ziqi-eval/blob/HEAD/src/five-shot.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"93be72957cb84414","mcp_get_code":{"code_sha256":"93be72957cb84414"}},{"arxiv_id":"2406.15885","paper":"/paper/the-music-maestro-or-the-musically-challenged","title":"The Music Maestro or The Musically Challenged, A Massive Music Evaluation Benchmark for Large Language Models","date":"2024-06-22","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"zcli-charlie/ziqi-eval","path":"src/zero-shot-eval.py","file_url":"https://github.com/zcli-charlie/ziqi-eval/blob/HEAD/src/zero-shot-eval.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"6b015217d6b3400c","mcp_get_code":{"code_sha256":"6b015217d6b3400c"}},{"arxiv_id":"ijcai2025_1031","paper":null,"title":"arXiv:ijcai2025_1031","date":null,"month_inferred_from_arxiv_id":null,"title_source":null,"repo":"yczju/cycseq","path":"dataset/preprocess.py","file_url":"https://github.com/yczju/cycseq/blob/HEAD/dataset/preprocess.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"c9f8d4041f56441a","mcp_get_code":{"code_sha256":"c9f8d4041f56441a"}}]}