{"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/collect-results","entry":"collect_results","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":0,"n_places":6,"n_places_pointer_only":3,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":0,"ran_fixture":0,"ran":2,"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":"2608.30768","paper":"/paper/arxiv-2608-30768","title":"CORAL: A Benchmark for Structure-aware and Brain-wide Neuron Reconstruction in Light Microscopy","date":null,"month_inferred_from_arxiv_id":"2026-08","title_source":"syntology","repo":"yang-ze-kang/CORAL","path":"calculate_metrics.py","file_url":"https://github.com/yang-ze-kang/CORAL/blob/HEAD/calculate_metrics.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"code_sha256_prefix":"f1604f2bf6b03522","mcp_get_code":{"code_sha256":"f1604f2bf6b03522"}},{"arxiv_id":"2605.11388","paper":"/paper/arxiv-2605-11388","title":"Deep Reasoning in General Purpose Agents via Structured Meta-Cognition","date":null,"month_inferred_from_arxiv_id":"2026-05","title_source":"syntology","repo":"DeanLight/dolores","path":"analysis/dr_analysis.py","file_url":"https://github.com/DeanLight/dolores/blob/HEAD/analysis/dr_analysis.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"cfeb5ccc75ffe9f9","mcp_get_code":{"code_sha256":"cfeb5ccc75ffe9f9"}},{"arxiv_id":"2505.21866","paper":"/paper/csi-bench-a-large-scale-in-the-wild-dataset","title":"CSI-Bench: A Large-Scale In-the-Wild Dataset for Multi-task WiFi Sensing","date":"2025-05-28","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"Jenny-Zhu/CSI-Bench-Real-WiFi-Sensing-Benchmark","path":"result_analysis/all_result_summary.py","file_url":"https://github.com/Jenny-Zhu/CSI-Bench-Real-WiFi-Sensing-Benchmark/blob/HEAD/result_analysis/all_result_summary.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"ba3123c0e55663e7","mcp_get_code":{"code_sha256":"ba3123c0e55663e7"}},{"arxiv_id":"2406.12624","paper":"/paper/judging-the-judges-evaluating-alignment-and","title":"Judging the Judges: Evaluating Alignment and Vulnerabilities in LLMs-as-Judges","date":"2024-06-18","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"UMass-Meta-LLM-Eval/llm_eval","path":"src/llm_eval/analysis/import_data.py","file_url":"https://github.com/UMass-Meta-LLM-Eval/llm_eval/blob/HEAD/src/llm_eval/analysis/import_data.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"c4cd0f2b59806fd9","mcp_get_code":{"code_sha256":"c4cd0f2b59806fd9"}},{"arxiv_id":"2401.17044","paper":"/paper/scalable-mechanism-design-for-multi-agent","title":"Scalable Mechanism Design for Multi-Agent Path Finding","date":"2024-01-30","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"lunjohnzhang/MAPF-Mechanism","path":"analysis/plot_stats_vs_n_agents.py","file_url":"https://github.com/lunjohnzhang/MAPF-Mechanism/blob/HEAD/analysis/plot_stats_vs_n_agents.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"7aaaa9025161e745","mcp_get_code":{"code_sha256":"7aaaa9025161e745"}},{"arxiv_id":"2311.07633","paper":"/paper/rethinking-and-benchmarking-predict-then","title":"Benchmarking PtO and PnO Methods in the Predictive Combinatorial Optimization Regime","date":"2023-11-13","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"Thinklab-SJTU/PredictiveCO-Benchmark","path":"rethink_exp/collect_results.py","file_url":"https://github.com/Thinklab-SJTU/PredictiveCO-Benchmark/blob/HEAD/rethink_exp/collect_results.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"dde3bb6970d2d582","mcp_get_code":{"code_sha256":"dde3bb6970d2d582"}}]}