{"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/preserving-commonsense-knowledge-from-pre","title":"Preserving Commonsense Knowledge from Pre-trained Language Models via Causal Inference","arxiv_id":"2306.10790","date":"2023-06-19","proceeding":null,"authors":["Junhao Zheng","Qianli Ma","Shengjie Qiu","Yue Wu","Peitian Ma","Junlong Liu","Huawen Feng","Xichen Shang","Haibin Chen"],"abstract":"Fine-tuning has been proven to be a simple and effective technique to transfer the learned knowledge of Pre-trained Language Models (PLMs) to downstream tasks. However, vanilla fine-tuning easily overfits the target data and degrades the generalization ability. Most existing studies attribute it to catastrophic forgetting, and they retain the pre-trained knowledge indiscriminately without identifying what knowledge is transferable. Motivated by this, we frame fine-tuning into a causal graph and discover that the crux of catastrophic forgetting lies in the missing causal effects from the pretrained data. Based on the causal view, we propose a unified objective for fine-tuning to retrieve the causality back. Intriguingly, the unified objective can be seen as the sum of the vanilla fine-tuning objective, which learns new knowledge from target data, and the causal objective, which preserves old knowledge from PLMs. Therefore, our method is flexible and can mitigate negative transfer while preserving knowledge. Since endowing models with commonsense is a long-standing challenge, we implement our method on commonsense QA with a proposed heuristic estimation to verify its effectiveness. In the experiments, our method outperforms state-of-the-art fine-tuning methods on all six commonsense QA datasets and can be implemented as a plug-in module to inflate the performance of existing QA models.","url_abs":"https://arxiv.org/abs/2306.10790v1","url_pdf":"https://arxiv.org/pdf/2306.10790v1.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":"preserving-commonsense-knowledge-from-pre","repo_url":"https://github.com/qianlima-lab/cet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"preserving-commonsense-knowledge-from-pre","repo_url":"https://github.com/zzz47zzz/cet","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"causal-inference","task_name":"Causal Inference"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2306.10790","atlas_url":"https://app.syntology.ai/?focus=2306.10790","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2306.10790"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/qianlima-lab/cet","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/zzz47zzz/cet","reach":null}],"summary":{"ran":1,"unverified":1},"by_repo_kind":{"official":{"samples":2,"ran":1,"repositories":2}},"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":2,"samples":[{"code_sha256_prefix":"296df2ff034a4c22","entry":"RecAdam","repo":"qianlima-lab/cet","repo_kind":"official","path":"utils/optimization_utils.py","file_url":"https://github.com/qianlima-lab/cet/blob/HEAD/utils/optimization_utils.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":"296df2ff034a4c22"}},{"code_sha256_prefix":"53210d198668b27d","entry":"Mixout","repo":"zzz47zzz/cet","repo_kind":"official","path":"utils/mixout_utils.py","file_url":"https://github.com/zzz47zzz/cet/blob/HEAD/utils/mixout_utils.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":"53210d198668b27d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}