{"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-runs","entry":"load_runs","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":11,"n_papers_ran":4,"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":10,"n_samples_ran":3,"n_samples_fingerprinted":1,"n_places":12,"n_places_pointer_only":1,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":2,"ran_fixture":0,"ran":1,"unverified":7},"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":"2608.12489","paper":"/paper/arxiv-2608-12489","title":"When Can You Trust Offline Evaluation of Equal-Cost Top-k Allocation? A Controlled, Reproducible Benchmark and Practitioner's Guide","date":null,"month_inferred_from_arxiv_id":"2026-08","title_source":"syntology","repo":"binshuangli/allocation-ope-bench","path":"src/allocation_ope_bench/analysis/misspec.py","file_url":"https://github.com/binshuangli/allocation-ope-bench/blob/HEAD/src/allocation_ope_bench/analysis/misspec.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"22f69fd6c8d729de","mcp_get_code":{"code_sha256":"22f69fd6c8d729de"}},{"arxiv_id":"2601.11499","paper":"/paper/arxiv-2601-11499","title":"On the Probability of First Success in Differential Evolution: Hazard Identities and Tail Bounds","date":null,"month_inferred_from_arxiv_id":"2026-01","title_source":"syntology","repo":"snenovgmailcom/lshade_hazard_project","path":"analysis/plot_quasi_morse.py","file_url":"https://github.com/snenovgmailcom/lshade_hazard_project/blob/HEAD/analysis/plot_quasi_morse.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"a9db344bcccc8706","mcp_get_code":{"code_sha256":"a9db344bcccc8706"}},{"arxiv_id":"2408.04667","paper":"/paper/llm-stability-a-detailed-analysis-with-some","title":"Non-Determinism of \"Deterministic\" LLM Settings","date":"2024-08-06","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"Comcast/llm-stability","path":"evaluate.py","file_url":"https://github.com/Comcast/llm-stability/blob/HEAD/evaluate.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"151e9739708eb70c","mcp_get_code":{"code_sha256":"151e9739708eb70c"}},{"arxiv_id":"2408.04667","paper":"/paper/llm-stability-a-detailed-analysis-with-some","title":"Non-Determinism of \"Deterministic\" LLM Settings","date":"2024-08-06","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"breckbaldwin/llm-stability","path":"evaluate.py","file_url":"https://github.com/breckbaldwin/llm-stability/blob/HEAD/evaluate.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":"2aa869fcf2a593c3","mcp_get_code":{"code_sha256":"2aa869fcf2a593c3"}},{"arxiv_id":"2210.11698","paper":"/paper/learning-robust-dynamics-through-variational","title":"Learning Robust Dynamics through Variational Sparse Gating","date":"2022-10-21","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"arnavkj1995/vsg","path":"plotting.py","file_url":"https://github.com/arnavkj1995/vsg/blob/HEAD/plotting.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"151b79dfa34b9b00","mcp_get_code":{"code_sha256":"151b79dfa34b9b00"}},{"arxiv_id":"2202.09481","paper":"/paper/transdreamer-reinforcement-learning-with-1","title":"TransDreamer: Reinforcement Learning with Transformer World Models","date":"2022-02-19","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"danijar/crafter","path":"analysis/common.py","file_url":"https://github.com/danijar/crafter/blob/HEAD/analysis/common.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"75a8133b273fb929","mcp_get_code":{"code_sha256":"75a8133b273fb929"}},{"arxiv_id":"2110.14565","paper":"/paper/dreamerpro-reconstruction-free-model-based-1","title":"DreamerPro: Reconstruction-Free Model-Based Reinforcement Learning with Prototypical Representations","date":"2021-10-27","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"fdeng18/dreamer-pro","path":"Dreamer/common/plot.py","file_url":"https://github.com/fdeng18/dreamer-pro/blob/HEAD/Dreamer/common/plot.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"4878c13e37288f19","mcp_get_code":{"code_sha256":"4878c13e37288f19"}},{"arxiv_id":"2108.08787","paper":"/paper/mr-tydi-a-multi-lingual-benchmark-for-dense","title":"Mr. TyDi: A Multi-lingual Benchmark for Dense Retrieval","date":"2021-08-19","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":null,"path":"","file_url":null,"status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"code_sha256_prefix":"8735894f4e3d26ec","mcp_get_code":{"code_sha256":"8735894f4e3d26ec"}},{"arxiv_id":"2106.07156","paper":"/paper/temporal-predictive-coding-for-model-based","title":"Temporal Predictive Coding For Model-Based Planning In Latent Space","date":"2021-06-14","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"VinAIResearch/TPC-tensorflow","path":"plotting.py","file_url":"https://github.com/VinAIResearch/TPC-tensorflow/blob/HEAD/plotting.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"e6df7a3762e59661","mcp_get_code":{"code_sha256":"e6df7a3762e59661"}},{"arxiv_id":"2010.02193","paper":"/paper/mastering-atari-with-discrete-world-models-1","title":"Mastering Atari with Discrete World Models","date":"2020-10-05","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"iglu-contest/iglu-builder-baseline-dreamer","path":"common/plot.py","file_url":"https://github.com/iglu-contest/iglu-builder-baseline-dreamer/blob/HEAD/common/plot.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"4878c13e37288f19","mcp_get_code":{"code_sha256":"4878c13e37288f19"}},{"arxiv_id":"2003.05002","paper":"/paper/tydi-qa-a-benchmark-for-information-seeking","title":"TyDi QA: A Benchmark for Information-Seeking Question Answering in Typologically Diverse Languages","date":"2020-03-10","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"castorini/mr.tydi","path":"scripts/hybrid.py","file_url":"https://github.com/castorini/mr.tydi/blob/HEAD/scripts/hybrid.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"8735894f4e3d26ec","mcp_get_code":{"code_sha256":"8735894f4e3d26ec"}},{"arxiv_id":"1912.01603","paper":"/paper/dream-to-control-learning-behaviors-by-latent","title":"Dream to Control: Learning Behaviors by Latent Imagination","date":"2019-12-03","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"danijar/dreamer","path":"plotting.py","file_url":"https://github.com/danijar/dreamer/blob/HEAD/plotting.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"6dd3eb82ea8db085","mcp_get_code":{"code_sha256":"6dd3eb82ea8db085"}}]}