{"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-eval-template","entry":"get_eval_template","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":0,"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":3,"n_samples_ran":0,"n_samples_fingerprinted":0,"n_places":7,"n_places_pointer_only":1,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":0,"ran_fixture":0,"ran":0,"unverified":3},"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":"2605.28822","paper":"/paper/arxiv-2605-28822","title":"Lightweight Multimodal LLM-Enabled Cost-Effective Defect Grading of Power Transmission Equipment","date":null,"month_inferred_from_arxiv_id":"2026-05","title_source":"syntology","repo":"hiyouga/LlamaFactory","path":"src/llamafactory/eval/template.py","file_url":"https://github.com/hiyouga/LlamaFactory/blob/HEAD/src/llamafactory/eval/template.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":"14abb4615516bc7e","mcp_get_code":{"code_sha256":"14abb4615516bc7e"}},{"arxiv_id":"2505.18672","paper":"/paper/does-representation-intervention-really","title":"Does Representation Intervention Really Identify Desired Concepts and Elicit Alignment?","date":"2025-05-24","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"tmlr-group/COCA","path":"src/llamafactory/eval/template.py","file_url":"https://github.com/tmlr-group/COCA/blob/HEAD/src/llamafactory/eval/template.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":"14abb4615516bc7e","mcp_get_code":{"code_sha256":"14abb4615516bc7e"}},{"arxiv_id":"2410.18640","paper":"/paper/weak-to-strong-preference-optimization","title":"Weak-to-Strong Preference Optimization: Stealing Reward from Weak Aligned Model","date":"2024-10-24","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"zwhong714/weak-to-strong-preference-optimization","path":"src/llamafactory/eval/template.py","file_url":"https://github.com/zwhong714/weak-to-strong-preference-optimization/blob/HEAD/src/llamafactory/eval/template.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":"81d83e0fa1ee57f3","mcp_get_code":{"code_sha256":"81d83e0fa1ee57f3"}},{"arxiv_id":"2410.07672","paper":"/paper/macpo-weak-to-strong-alignment-via-multi","title":"MACPO: Weak-to-Strong Alignment via Multi-Agent Contrastive Preference Optimization","date":"2024-10-10","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"youganglyu/MACPO","path":"src/llmtuner/eval/template.py","file_url":"https://github.com/youganglyu/MACPO/blob/HEAD/src/llmtuner/eval/template.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"81d83e0fa1ee57f3","mcp_get_code":{"code_sha256":"81d83e0fa1ee57f3"}},{"arxiv_id":"2405.19266","paper":"/paper/pediatricsgpt-large-language-models-as","title":"PediatricsGPT: Large Language Models as Chinese Medical Assistants for Pediatric Applications","date":"2024-05-29","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"ydk122024/PediatricsGPT","path":"src/llmtuner/eval/template.py","file_url":"https://github.com/ydk122024/PediatricsGPT/blob/HEAD/src/llmtuner/eval/template.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"81d83e0fa1ee57f3","mcp_get_code":{"code_sha256":"81d83e0fa1ee57f3"}},{"arxiv_id":"2403.13786","paper":"/paper/chain-of-interaction-enhancing-large-language","title":"Chain-of-Interaction: Enhancing Large Language Models for Psychiatric Behavior Understanding by Dyadic Contexts","date":"2024-03-20","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"trust-nlp/coi-psychotherapy","path":"src/llmtuner/eval/MMCBtemplate.py","file_url":"https://github.com/trust-nlp/coi-psychotherapy/blob/HEAD/src/llmtuner/eval/MMCBtemplate.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":"e180bbd09fe3a2f2","mcp_get_code":{"code_sha256":"e180bbd09fe3a2f2"}},{"arxiv_id":"aaai_34615","paper":null,"title":"arXiv:aaai_34615","date":null,"month_inferred_from_arxiv_id":null,"title_source":null,"repo":"thaihungle/MRPO","path":"src/llmtuner/eval/template.py","file_url":"https://github.com/thaihungle/MRPO/blob/HEAD/src/llmtuner/eval/template.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":"81d83e0fa1ee57f3","mcp_get_code":{"code_sha256":"81d83e0fa1ee57f3"}}]}