{"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/normalize-scores","entry":"normalize_scores","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":8,"n_papers_ran":3,"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":9,"n_samples_ran":3,"n_samples_fingerprinted":2,"n_places":9,"n_places_pointer_only":4,"by_status":{"ran_honours":0,"ran_violates":1,"ran_draft_wrong":0,"ran_fixture":1,"ran":1,"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":"2605.28062","paper":"/paper/arxiv-2605-28062","title":"ConvMemory: A Lightweight Learned Memory Reranker, a Negative Attribution Result, and a Research-Preview Conflict Editor","date":null,"month_inferred_from_arxiv_id":"2026-05","title_source":"syntology","repo":"pth2002/ConvMemory","path":"convmemory/reranker.py","file_url":"https://github.com/pth2002/ConvMemory/blob/HEAD/convmemory/reranker.py","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"ee33c4c15d79bb72","mcp_get_code":{"code_sha256":"ee33c4c15d79bb72"}},{"arxiv_id":"2602.05786","paper":"/paper/arxiv-2602-05786","title":"Selecting Hyperparameters for Tree-Boosting","date":null,"month_inferred_from_arxiv_id":"2026-02","title_source":"syntology","repo":"fl0risk/HPOTreeBoosting","path":"utils_plot_maker.py","file_url":"https://github.com/fl0risk/HPOTreeBoosting/blob/HEAD/utils_plot_maker.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"743b09f8b88855bf","mcp_get_code":{"code_sha256":"743b09f8b88855bf"}},{"arxiv_id":"2602.05786","paper":"/paper/arxiv-2602-05786","title":"Selecting Hyperparameters for Tree-Boosting","date":null,"month_inferred_from_arxiv_id":"2026-02","title_source":"syntology","repo":"fl0risk/HPOTreeBoosting","path":"utils_plots_comparison.py","file_url":"https://github.com/fl0risk/HPOTreeBoosting/blob/HEAD/utils_plots_comparison.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"84816301a3b34104","mcp_get_code":{"code_sha256":"84816301a3b34104"}},{"arxiv_id":"2502.14911","paper":null,"title":"arXiv:2502.14911","date":null,"month_inferred_from_arxiv_id":"2025-02","title_source":null,"repo":"aisingapore/sea-helm","path":"src/aggregation/bootstrap.py","file_url":"https://github.com/aisingapore/sea-helm/blob/HEAD/src/aggregation/bootstrap.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"05c5b44e383af8f1","mcp_get_code":{"code_sha256":"05c5b44e383af8f1"}},{"arxiv_id":"2403.02839","paper":"/paper/an-empirical-study-of-llm-as-a-judge-for-llm","title":"An Empirical Study of LLM-as-a-Judge for LLM Evaluation: Fine-tuned Judge Model is not a General Substitute for GPT-4","date":"2024-03-05","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"huihuichyan/unlimitedjudge","path":"src/evaluate_reliability.py","file_url":"https://github.com/huihuichyan/unlimitedjudge/blob/HEAD/src/evaluate_reliability.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"972a8f7586d01eca","mcp_get_code":{"code_sha256":"972a8f7586d01eca"}},{"arxiv_id":"2305.13829","paper":"/paper/learn-from-mistakes-through-cooperative","title":"Learning from Mistakes via Cooperative Study Assistant for Large Language Models","date":"2023-05-23","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"madaan/self-refine","path":"src/acronym/run_mcts.py","file_url":"https://github.com/madaan/self-refine/blob/HEAD/src/acronym/run_mcts.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":"94f412dc2125cdd8","mcp_get_code":{"code_sha256":"94f412dc2125cdd8"}},{"arxiv_id":"1705.02304","paper":"/paper/deep-speaker-an-end-to-end-neural-speaker","title":"Deep Speaker: an End-to-End Neural Speaker Embedding System","date":"2017-05-05","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"Walleclipse/Deep_Speaker-speaker_recognition_system","path":"test_model.py","file_url":"https://github.com/Walleclipse/Deep_Speaker-speaker_recognition_system/blob/HEAD/test_model.py","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"6a6418cda28d6455","mcp_get_code":{"code_sha256":"6a6418cda28d6455"}},{"arxiv_id":"openreview_BgNkAHzveU","paper":null,"title":"arXiv:openreview_BgNkAHzveU","date":null,"month_inferred_from_arxiv_id":null,"title_source":null,"repo":"ph-phuc/MetaEns","path":"utils/evaluation.py","file_url":"https://github.com/ph-phuc/MetaEns/blob/HEAD/utils/evaluation.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"1b0b0144264029e7","mcp_get_code":{"code_sha256":"1b0b0144264029e7"}},{"arxiv_id":"2022.emnlp-main.727","paper":null,"title":"arXiv:2022.emnlp-main.727","date":null,"month_inferred_from_arxiv_id":null,"title_source":null,"repo":"gzerveas/CODER","path":"convert_file_format.py","file_url":"https://github.com/gzerveas/CODER/blob/HEAD/convert_file_format.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"code_sha256_prefix":"d315e0701b749fbc","mcp_get_code":{"code_sha256":"d315e0701b749fbc"}}]}