{"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/get-questions","entry":"get_questions","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":5,"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":4,"n_samples_ran":3,"n_samples_fingerprinted":0,"n_places":6,"n_places_pointer_only":1,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":0,"ran_fixture":0,"ran":3,"unverified":1},"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":"2605.20075","paper":"/paper/arxiv-2605-20075","title":"CopT: Contrastive On-Policy Thinking with Continuous Spaces for General and Agentic Reasoning","date":null,"month_inferred_from_arxiv_id":"2026-05","title_source":"syntology","repo":"shishirpatil/gorilla","path":"gorilla/inference/gorilla_eval.py","file_url":"https://github.com/shishirpatil/gorilla/blob/HEAD/gorilla/inference/gorilla_eval.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"899cb7d14116e83d","mcp_get_code":{"code_sha256":"899cb7d14116e83d"}},{"arxiv_id":"2406.13356","paper":"/paper/jogging-the-memory-of-unlearned-model-through","title":"Jogging the Memory of Unlearned LLMs Through Targeted Relearning Attacks","date":"2024-06-19","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"s-huu/jog_llm_memory","path":"wmdp/src/gpt4_eval.py","file_url":"https://github.com/s-huu/jog_llm_memory/blob/HEAD/wmdp/src/gpt4_eval.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"c85b881c2039a329","mcp_get_code":{"code_sha256":"c85b881c2039a329"}},{"arxiv_id":"2406.09948","paper":"/paper/blend-a-benchmark-for-llms-on-everyday","title":"BLEnD: A Benchmark for LLMs on Everyday Knowledge in Diverse Cultures and Languages","date":"2024-06-14","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"nlee0212/blend","path":"evaluation/evaluation_utils.py","file_url":"https://github.com/nlee0212/blend/blob/HEAD/evaluation/evaluation_utils.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"ad8827896937a543","mcp_get_code":{"code_sha256":"ad8827896937a543"}},{"arxiv_id":"2404.08700","paper":"/paper/is-your-llm-outdated-benchmarking-llms","title":"DyKnow: Dynamically Verifying Time-Sensitive Factual Knowledge in LLMs","date":"2024-04-10","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"sislab-unitn/dyknow","path":"models_output/utils.py","file_url":"https://github.com/sislab-unitn/dyknow/blob/HEAD/models_output/utils.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"f48962cb47272ac5","mcp_get_code":{"code_sha256":"f48962cb47272ac5"}},{"arxiv_id":"2404.06921","paper":"/paper/goex-perspectives-and-designs-towards-a","title":"GoEX: Perspectives and Designs Towards a Runtime for Autonomous LLM Applications","date":"2024-04-10","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"ShishirPatil/gorilla","path":"gorilla/inference/gorilla_eval.py","file_url":"https://github.com/ShishirPatil/gorilla/blob/HEAD/gorilla/inference/gorilla_eval.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"code_sha256_prefix":"899cb7d14116e83d","mcp_get_code":{"code_sha256":"899cb7d14116e83d"}},{"arxiv_id":"2401.10727","paper":"/paper/tool-lmm-a-large-multi-modal-model-for-tool","title":"MLLM-Tool: A Multimodal Large Language Model For Tool Agent Learning","date":"2024-01-19","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"mllm-tool/mllm-tool","path":"code/inference.py","file_url":"https://github.com/mllm-tool/mllm-tool/blob/HEAD/code/inference.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"899cb7d14116e83d","mcp_get_code":{"code_sha256":"899cb7d14116e83d"}}]}