{"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-cases","entry":"load_cases","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":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":6,"n_samples_ran":1,"n_samples_fingerprinted":0,"n_places":6,"n_places_pointer_only":0,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":0,"ran_fixture":0,"ran":1,"unverified":5},"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":"2609.04105","paper":"/paper/arxiv-2609-04105","title":"Hardware-Aware FP4 FlashAttention-4","date":null,"month_inferred_from_arxiv_id":"2026-09","title_source":"syntology","repo":"MrHuff/fp4-fa4","path":"results/fp4_fa4_unified_20260801/build_summary.py","file_url":"https://github.com/MrHuff/fp4-fa4/blob/HEAD/results/fp4_fa4_unified_20260801/build_summary.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":"f50de98ba6a7a99b","mcp_get_code":{"code_sha256":"f50de98ba6a7a99b"}},{"arxiv_id":"2607.24878","paper":"/paper/arxiv-2607-24878","title":"GraphRareBench: An Auditable Graph-Evidence Benchmark for Phenotype-Driven Rare-Disease Diagnosis","date":null,"month_inferred_from_arxiv_id":"2026-07","title_source":"syntology","repo":"GUI0609/GraphRareBench","path":"src/graphrarebench/io.py","file_url":"https://github.com/GUI0609/GraphRareBench/blob/HEAD/src/graphrarebench/io.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"799b98203edcd04d","mcp_get_code":{"code_sha256":"799b98203edcd04d"}},{"arxiv_id":"2602.15898","paper":"/paper/arxiv-2602-15898","title":"MultiCube-RAG for Multi-hop Question Answering","date":null,"month_inferred_from_arxiv_id":"2026-02","title_source":"syntology","repo":"OpenBMB/UltraRAG","path":"script/case_study.py","file_url":"https://github.com/OpenBMB/UltraRAG/blob/HEAD/script/case_study.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":"8ef4d0e3d5ff2a82","mcp_get_code":{"code_sha256":"8ef4d0e3d5ff2a82"}},{"arxiv_id":"2401.04621","paper":"/paper/debugbench-evaluating-debugging-capability-of","title":"DebugBench: Evaluating Debugging Capability of Large Language Models","date":"2024-01-09","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"thunlp/debugbench","path":"dataset_construction/generate_multiples.py","file_url":"https://github.com/thunlp/debugbench/blob/HEAD/dataset_construction/generate_multiples.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":"140a272ea2caef78","mcp_get_code":{"code_sha256":"140a272ea2caef78"}},{"arxiv_id":"2401.04621","paper":"/paper/debugbench-evaluating-debugging-capability-of","title":"DebugBench: Evaluating Debugging Capability of Large Language Models","date":"2024-01-09","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"thunlp/debugbench","path":"dataset_construction/generate_singles.py","file_url":"https://github.com/thunlp/debugbench/blob/HEAD/dataset_construction/generate_singles.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":"8a7b54cf614e7e35","mcp_get_code":{"code_sha256":"8a7b54cf614e7e35"}},{"arxiv_id":"2311.03415","paper":"/paper/powerflownet-leveraging-message-passing-gnns","title":"PowerFlowNet: Power Flow Approximation Using Message Passing Graph Neural Networks","date":"2023-11-06","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"StavrosOrf/PoweFlowNet","path":"speedup_evaluator.py","file_url":"https://github.com/StavrosOrf/PoweFlowNet/blob/HEAD/speedup_evaluator.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"981325d5db79c1c4","mcp_get_code":{"code_sha256":"981325d5db79c1c4"}}]}