{"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/find-pruneable-heads-and-indices","entry":"find_pruneable_heads_and_indices","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":19,"n_papers_ran":14,"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":12,"n_samples_ran":7,"n_samples_fingerprinted":0,"n_places":19,"n_places_pointer_only":7,"by_status":{"ran_honours":0,"ran_violates":0,"ran_draft_wrong":4,"ran_fixture":0,"ran":3,"unverified":5},"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":"2607.02956","paper":"/paper/arxiv-2607-02956","title":"MORE: A Multilingual Document Parsing Benchmark and Evaluation","date":null,"month_inferred_from_arxiv_id":"2026-07","title_source":"syntology","repo":"datalab-to/surya","path":"surya/common/compat.py","file_url":"https://github.com/datalab-to/surya/blob/HEAD/surya/common/compat.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":"cb0ea0c51176639b","mcp_get_code":{"code_sha256":"cb0ea0c51176639b"}},{"arxiv_id":"2602.22581","paper":"/paper/arxiv-2602-22581","title":"IBCircuit: Towards Holistic Circuit Discovery with Information Bottleneck","date":null,"month_inferred_from_arxiv_id":"2026-02","title_source":"syntology","repo":"ivanniu/IBCircuit","path":"models/gpt2_IB.py","file_url":"https://github.com/ivanniu/IBCircuit/blob/HEAD/models/gpt2_IB.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"86ce14ff7ea60630","mcp_get_code":{"code_sha256":"86ce14ff7ea60630"}},{"arxiv_id":"2405.10974","paper":"/paper/bottleneck-minimal-indexing-for-generative","title":"Bottleneck-Minimal Indexing for Generative Document Retrieval","date":"2024-05-12","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"kduxin/Bottleneck-Minimal-Indexing","path":"NCIRetriever/nci_transformers/modeling_utils.py","file_url":"https://github.com/kduxin/Bottleneck-Minimal-Indexing/blob/HEAD/NCIRetriever/nci_transformers/modeling_utils.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"6d254a1b39c650ff","mcp_get_code":{"code_sha256":"6d254a1b39c650ff"}},{"arxiv_id":"2402.13040","paper":"/paper/text-guided-molecule-generation-with","title":"Text-Guided Molecule Generation with Diffusion Language Model","date":"2024-02-20","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"Deno-V/tgm-dlm","path":"improved-diffusion/improved_diffusion/transformer_model.py","file_url":"https://github.com/Deno-V/tgm-dlm/blob/HEAD/improved-diffusion/improved_diffusion/transformer_model.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"ffe4fa9032a930ff","mcp_get_code":{"code_sha256":"ffe4fa9032a930ff"}},{"arxiv_id":"2312.12470","paper":"/paper/rotated-multi-scale-interaction-network-for","title":"Rotated Multi-Scale Interaction Network for Referring Remote Sensing Image Segmentation","date":"2023-12-19","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"lsan2401/rmsin","path":"bert/modeling_utils.py","file_url":"https://github.com/lsan2401/rmsin/blob/HEAD/bert/modeling_utils.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"b239247b87d4f3dc","mcp_get_code":{"code_sha256":"b239247b87d4f3dc"}},{"arxiv_id":"2312.12198","paper":"/paper/mask-grounding-for-referring-image","title":"Mask Grounding for Referring Image Segmentation","date":"2023-12-19","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"yxchng/mask-grounding","path":"bert/modeling_utils.py","file_url":"https://github.com/yxchng/mask-grounding/blob/HEAD/bert/modeling_utils.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"AGPL-3.0","inline_ok":false,"code_sha256_prefix":"b239247b87d4f3dc","mcp_get_code":{"code_sha256":"b239247b87d4f3dc"}},{"arxiv_id":"2309.01017","paper":"/paper/contrastive-grouping-with-transformer-for-1","title":"Contrastive Grouping with Transformer for Referring Image Segmentation","date":"2023-09-02","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"toneyaya/cgformer","path":"bert/modeling_utils.py","file_url":"https://github.com/toneyaya/cgformer/blob/HEAD/bert/modeling_utils.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"b239247b87d4f3dc","mcp_get_code":{"code_sha256":"b239247b87d4f3dc"}},{"arxiv_id":"2308.13853","paper":"/paper/beyond-one-to-one-rethinking-the-referring","title":"Beyond One-to-One: Rethinking the Referring Image Segmentation","date":"2023-08-26","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"toggle1995/RIS-DMMI","path":"bert/modeling_utils.py","file_url":"https://github.com/toggle1995/RIS-DMMI/blob/HEAD/bert/modeling_utils.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"b239247b87d4f3dc","mcp_get_code":{"code_sha256":"b239247b87d4f3dc"}},{"arxiv_id":"2303.14582","paper":"/paper/identification-of-negative-transfers-in","title":"Identification of Negative Transfers in Multitask Learning Using Surrogate Models","date":"2023-03-25","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"neu-statsml-research/task-modeling","path":"transformers/modeling_utils.py","file_url":"https://github.com/neu-statsml-research/task-modeling/blob/HEAD/transformers/modeling_utils.py","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"6d254a1b39c650ff","mcp_get_code":{"code_sha256":"6d254a1b39c650ff"}},{"arxiv_id":"2211.11694","paper":"/paper/exploring-discrete-diffusion-models-for-image","title":"Exploring Discrete Diffusion Models for Image Captioning","date":"2022-11-21","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"buxiangzhiren/ddcap","path":"fix_pre.py","file_url":"https://github.com/buxiangzhiren/ddcap/blob/HEAD/fix_pre.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"adea8f61592be69f","mcp_get_code":{"code_sha256":"adea8f61592be69f"}},{"arxiv_id":"2110.00855","paper":"/paper/survtrace-transformers-for-survival-analysis","title":"SurvTRACE: Transformers for Survival Analysis with Competing Events","date":"2021-10-02","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"RyanWangZf/SurvTRACE","path":"survtrace/modeling_bert.py","file_url":"https://github.com/RyanWangZf/SurvTRACE/blob/HEAD/survtrace/modeling_bert.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"26b7e24933ef7cd5","mcp_get_code":{"code_sha256":"26b7e24933ef7cd5"}},{"arxiv_id":"2109.03808","paper":"/paper/smelting-gold-and-silver-for-improved","title":"Smelting Gold and Silver for Improved Multilingual AMR-to-Text Generation","date":"2021-09-08","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"UKPLab/m-AMR2Text","path":"transformers/modeling_utils.py","file_url":"https://github.com/UKPLab/m-AMR2Text/blob/HEAD/transformers/modeling_utils.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":"6d254a1b39c650ff","mcp_get_code":{"code_sha256":"6d254a1b39c650ff"}},{"arxiv_id":"2108.13587","paper":"/paper/t3-vis-a-visual-analytic-framework-for","title":"T3-Vis: a visual analytic framework for Training and fine-Tuning Transformers in NLP","date":"2021-08-31","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"raymondzmc/T3-Vis","path":"application/utils/head_importance.py","file_url":"https://github.com/raymondzmc/T3-Vis/blob/HEAD/application/utils/head_importance.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"code_sha256_prefix":"fb962beb65f3f102","mcp_get_code":{"code_sha256":"fb962beb65f3f102"}},{"arxiv_id":"2108.10904","paper":"/paper/simvlm-simple-visual-language-model","title":"SimVLM: Simple Visual Language Model Pretraining with Weak Supervision","date":"2021-08-24","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"FerryHuang/SimVLM","path":"simvlm/modeling_simvlm.py","file_url":"https://github.com/FerryHuang/SimVLM/blob/HEAD/simvlm/modeling_simvlm.py","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"code_sha256_prefix":"a98857f046bfe3d9","mcp_get_code":{"code_sha256":"a98857f046bfe3d9"}},{"arxiv_id":"2105.15203","paper":"/paper/segformer-simple-and-efficient-design-for","title":"SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers","date":"2021-05-31","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"VikParuchuri/surya","path":"surya/common/compat.py","file_url":"https://github.com/VikParuchuri/surya/blob/HEAD/surya/common/compat.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":"cb0ea0c51176639b","mcp_get_code":{"code_sha256":"cb0ea0c51176639b"}},{"arxiv_id":"2104.08066","paper":"/paper/effect-of-vision-and-language-extensions-on","title":"Effect of Visual Extensions on Natural Language Understanding in Vision-and-Language Models","date":"2021-04-16","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"alab-nii/eval_vl_glue","path":"eval_vl_glue/transformers_volta/modeling_utils.py","file_url":"https://github.com/alab-nii/eval_vl_glue/blob/HEAD/eval_vl_glue/transformers_volta/modeling_utils.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":"6d254a1b39c650ff","mcp_get_code":{"code_sha256":"6d254a1b39c650ff"}},{"arxiv_id":"2004.10964","paper":"/paper/don-t-stop-pretraining-adapt-language-models","title":"Don't Stop Pretraining: Adapt Language Models to Domains and Tasks","date":"2020-04-23","month_inferred_from_arxiv_id":null,"title_source":"archive","repo":"shizhediao/t-dna","path":"TDNA/modeling.py","file_url":"https://github.com/shizhediao/t-dna/blob/HEAD/TDNA/modeling.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":"1b4bc5e6ef0e9288","mcp_get_code":{"code_sha256":"1b4bc5e6ef0e9288"}},{"arxiv_id":"2023.findings-acl.765","paper":null,"title":"arXiv:2023.findings-acl.765","date":null,"month_inferred_from_arxiv_id":null,"title_source":null,"repo":"qtli/EIB","path":"code/EIB_model/modeling_gpt2.py","file_url":"https://github.com/qtli/EIB/blob/HEAD/code/EIB_model/modeling_gpt2.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":"510183a9fa81eacb","mcp_get_code":{"code_sha256":"510183a9fa81eacb"}},{"arxiv_id":"2021.emnlp-main.154","paper":null,"title":"arXiv:2021.emnlp-main.154","date":null,"month_inferred_from_arxiv_id":null,"title_source":null,"repo":"Hazelsuko07/TextHide","path":"transformers_hide/modeling_utils.py","file_url":"https://github.com/Hazelsuko07/TextHide/blob/HEAD/transformers_hide/modeling_utils.py","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"code_sha256_prefix":"02fd40244433a6e6","mcp_get_code":{"code_sha256":"02fd40244433a6e6"}}]}