{"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":"/paper/knowledge-preserving-pruning-for-pre-trained","title":"Accurate Retraining-free Pruning for Pretrained Encoder-based Language Models","arxiv_id":"2308.03449","date":"2023-08-07","proceeding":null,"authors":["Seungcheol Park","Hojun Choi","U Kang"],"abstract":"Given a pretrained encoder-based language model, how can we accurately compress it without retraining? Retraining-free structured pruning algorithms are crucial in pretrained language model compression due to their significantly reduced pruning cost and capability to prune large language models. However, existing retraining-free algorithms encounter severe accuracy degradation, as they fail to handle pruning errors, especially at high compression rates. In this paper, we propose K-prune (Knowledge-preserving pruning), an accurate retraining-free structured pruning algorithm for pretrained encoder-based language models. K-prune focuses on preserving the useful knowledge of the pretrained model to minimize pruning errors through a carefully designed iterative pruning process composed of knowledge measurement, knowledge-preserving mask search, and knowledge-preserving weight-tuning. As a result, K-prune shows significant accuracy improvements up to 58.02%p higher F1 score compared to existing retraining-free pruning algorithms under a high compression rate of 80% on the SQuAD benchmark without any retraining process.","url_abs":"https://arxiv.org/abs/2308.03449v2","url_pdf":"https://arxiv.org/pdf/2308.03449v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"knowledge-preserving-pruning-for-pre-trained","repo_url":"https://github.com/snudm-starlab/k-prune","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"model-compression","task_name":"Model Compression"}],"methods":[{"method_slug":"pruning","method_name":"Pruning"},{"method_slug":"fail","method_name":"fail"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2308.03449","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2308.03449"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/snudm-starlab/k-prune","reach":{"status":"ok"}}],"summary":{"ran":12,"unverified":4},"by_repo_kind":{"official":{"samples":16,"ran":12,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":16,"samples":[{"code_sha256_prefix":"81a976f2d07691ea","entry":"avg_seq_length","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/dataset/glue.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/dataset/glue.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"81a976f2d07691ea"}},{"code_sha256_prefix":"3099ee2eb84029fc","entry":"eval_squad_loss","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/evaluate/squad.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/evaluate/squad.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3099ee2eb84029fc"}},{"code_sha256_prefix":"9d9c2a0349979318","entry":"find_mask","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/prune.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/prune.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9d9c2a0349979318"}},{"code_sha256_prefix":"234ee0583e6aac86","entry":"flops_per_head","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/utils/capacity.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/utils/capacity.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"234ee0583e6aac86"}},{"code_sha256_prefix":"d931d90b153fe3b9","entry":"flops_per_neuron","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/utils/capacity.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/utils/capacity.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d931d90b153fe3b9"}},{"code_sha256_prefix":"7b5ef8fb4fcbf0d1","entry":"get_basemodel","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/utils/arch.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/utils/arch.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7b5ef8fb4fcbf0d1"}},{"code_sha256_prefix":"04b8921a51f5f4bc","entry":"get_encoder","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/utils/arch.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/utils/arch.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"04b8921a51f5f4bc"}},{"code_sha256_prefix":"4b6648bada434292","entry":"get_flops","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/utils/capacity.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/utils/capacity.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"4b6648bada434292"}},{"code_sha256_prefix":"e710f0bc27cebd18","entry":"get_layers","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/utils/arch.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/utils/arch.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"e710f0bc27cebd18"}},{"code_sha256_prefix":"a5972c71ba2fe6a0","entry":"max_seq_length","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/dataset/glue.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/dataset/glue.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a5972c71ba2fe6a0"}},{"code_sha256_prefix":"6d66e5c7114f28e9","entry":"num_labels","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/dataset/glue.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/dataset/glue.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6d66e5c7114f28e9"}},{"code_sha256_prefix":"0996b8ad517f85ab","entry":"prepare_validation_features","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/dataset/squad.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/dataset/squad.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0996b8ad517f85ab"}},{"code_sha256_prefix":"6cc4ed90b06ca999","entry":"compute_scores","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/prune.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/prune.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6cc4ed90b06ca999"}},{"code_sha256_prefix":"eb1a415db932da8d","entry":"kprune","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/prune.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/prune.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"eb1a415db932da8d"}},{"code_sha256_prefix":"a09579d64a06ab13","entry":"prepare_train_features","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/dataset/squad.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/dataset/squad.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a09579d64a06ab13"}},{"code_sha256_prefix":"62cdd999358b2ea8","entry":"squad_dataset","repo":"snudm-starlab/k-prune","repo_kind":"official","path":"src/dataset/squad.py","file_url":"https://github.com/snudm-starlab/k-prune/blob/HEAD/src/dataset/squad.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"62cdd999358b2ea8"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}