{"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/extract-response","entry":"extract_response","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":5,"n_samples_ran":1,"n_samples_fingerprinted":1,"n_places":7,"n_places_pointer_only":3,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":1,"ran_fixture":0,"ran":0,"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":"2604.16349","paper":"/paper/arxiv-2604-16349","title":"Benchmarking Real-Time Question Answering via Executable Code Workflows","date":null,"month_inferred_from_arxiv_id":"2026-04","title_source":"syntology","repo":"leaves-slient/RT-Bench","path":"DeepResearch/evaluation/evaluate_hle_official.py","file_url":"https://github.com/leaves-slient/RT-Bench/blob/HEAD/DeepResearch/evaluation/evaluate_hle_official.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"a4ee90a1bd5be041","mcp_get_code":{"code_sha256":"a4ee90a1bd5be041"}},{"arxiv_id":"2604.04017","paper":"/paper/arxiv-2604-04017","title":"GeoBrowse: A Geolocation Benchmark for Agentic Tool Use with Expert-Annotated Reasoning Traces","date":null,"month_inferred_from_arxiv_id":"2026-04","title_source":"syntology","repo":"ornamentt/GeoBrowse","path":"Evaluation/baseline/evaluate.py","file_url":"https://github.com/ornamentt/GeoBrowse/blob/HEAD/Evaluation/baseline/evaluate.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"a4ee90a1bd5be041","mcp_get_code":{"code_sha256":"a4ee90a1bd5be041"}},{"arxiv_id":"2603.09403","paper":"/paper/arxiv-2603-09403","title":"LLM as a Meta-Judge: Synthetic Data for NLP Evaluation Metric Validation","date":null,"month_inferred_from_arxiv_id":"2026-03","title_source":"syntology","repo":"eiglerl/meta-judge","path":"src/finetuning/train_grpo.py","file_url":"https://github.com/eiglerl/meta-judge/blob/HEAD/src/finetuning/train_grpo.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"5f9303d1a6f27bd6","mcp_get_code":{"code_sha256":"5f9303d1a6f27bd6"}},{"arxiv_id":"2509.25140","paper":"/paper/arxiv-2509-25140","title":"Re as on i ng Ban k: Scaling Agent Self-Evolving with Reasoning Memory","date":null,"month_inferred_from_arxiv_id":"2025-09","title_source":"syntology","repo":"google-research/reasoning-bank","path":"WebArena/autoeval/evaluate_trajectory.py","file_url":"https://github.com/google-research/reasoning-bank/blob/HEAD/WebArena/autoeval/evaluate_trajectory.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":"5503345bf4b138db","mcp_get_code":{"code_sha256":"5503345bf4b138db"}},{"arxiv_id":"2507.02592","paper":"/paper/websailor-navigating-super-human-reasoning","title":"WebSailor: Navigating Super-human Reasoning for Web Agent","date":"2025-07-03","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"alibaba-nlp/webwalker","path":"evaluation/evaluate_hle_official.py","file_url":"https://github.com/alibaba-nlp/webwalker/blob/HEAD/evaluation/evaluate_hle_official.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":"a4ee90a1bd5be041","mcp_get_code":{"code_sha256":"a4ee90a1bd5be041"}},{"arxiv_id":"2501.00316","paper":"/paper/mapeval-a-map-based-evaluation-of-geo-spatial","title":"MapEval: A Map-Based Evaluation of Geo-Spatial Reasoning in Foundation Models","date":"2024-12-31","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"MapEval/MapEval-Visual","path":"evaluation.py","file_url":"https://github.com/MapEval/MapEval-Visual/blob/HEAD/evaluation.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"402a187397a37a7c","mcp_get_code":{"code_sha256":"402a187397a37a7c"}},{"arxiv_id":"2024.findings-acl.8","paper":null,"title":"arXiv:2024.findings-acl.8","date":null,"month_inferred_from_arxiv_id":null,"title_source":null,"repo":"allenai/chime","path":"chime/src/utils.py","file_url":"https://github.com/allenai/chime/blob/HEAD/chime/src/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":"08c9f09c4d28a91c","mcp_get_code":{"code_sha256":"08c9f09c4d28a91c"}}]}