{"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/understanding-catastrophic-forgetting-in","title":"Understanding Catastrophic Forgetting in Language Models via Implicit Inference","arxiv_id":"2309.10105","date":"2023-09-18","proceeding":null,"authors":["Suhas Kotha","Jacob Mitchell Springer","aditi raghunathan"],"abstract":"We lack a systematic understanding of the effects of fine-tuning (via methods such as instruction-tuning or reinforcement learning from human feedback), particularly on tasks outside the narrow fine-tuning distribution. In a simplified scenario, we demonstrate that improving performance on tasks within the fine-tuning data distribution comes at the expense of capabilities on other tasks. We hypothesize that language models implicitly infer the task of the prompt and that fine-tuning skews this inference towards tasks in the fine-tuning distribution. To test this, we propose Conjugate Prompting, which artificially makes the task look farther from the fine-tuning distribution while requiring the same capability, and we find that this recovers some of the pretraining capabilities in our synthetic setup. Since real-world fine-tuning distributions are predominantly English, we apply conjugate prompting to recover pretrained capabilities in LLMs by simply translating the prompts to different languages. This allows us to recover in-context learning abilities lost via instruction tuning, natural reasoning capability lost during code fine-tuning, and, more concerningly, harmful content generation suppressed by safety fine-tuning in chatbots like ChatGPT.","url_abs":"https://arxiv.org/abs/2309.10105v2","url_pdf":"https://arxiv.org/pdf/2309.10105v2.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":"understanding-catastrophic-forgetting-in","repo_url":"https://github.com/kothasuhas/understanding-forgetting","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"in-context-learning","task_name":"In-Context Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2309.10105","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2309.10105"}},"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/kothasuhas/understanding-forgetting","reach":{"status":"ok"}}],"summary":{"ran":2,"unverified":4},"by_repo_kind":{"official":{"samples":6,"ran":2,"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":6,"samples":[{"code_sha256_prefix":"4ad874ded80c869c","entry":"format_mean_std","repo":"kothasuhas/understanding-forgetting","repo_kind":"official","path":"icl_vs_if/latex_tables.py","file_url":"https://github.com/kothasuhas/understanding-forgetting/blob/HEAD/icl_vs_if/latex_tables.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":"4ad874ded80c869c"}},{"code_sha256_prefix":"e6bd952ec83310a7","entry":"get_results_dict","repo":"kothasuhas/understanding-forgetting","repo_kind":"official","path":"icl_vs_if/evaluate_accuracy.py","file_url":"https://github.com/kothasuhas/understanding-forgetting/blob/HEAD/icl_vs_if/evaluate_accuracy.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":"e6bd952ec83310a7"}},{"code_sha256_prefix":"17e1b5f8edda1ed3","entry":"get_latex_cell","repo":"kothasuhas/understanding-forgetting","repo_kind":"official","path":"icl_vs_if/latex_tables.py","file_url":"https://github.com/kothasuhas/understanding-forgetting/blob/HEAD/icl_vs_if/latex_tables.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":"17e1b5f8edda1ed3"}},{"code_sha256_prefix":"bd78cc974c9ed25e","entry":"get_mean_std","repo":"kothasuhas/understanding-forgetting","repo_kind":"official","path":"icl_vs_if/latex_tables.py","file_url":"https://github.com/kothasuhas/understanding-forgetting/blob/HEAD/icl_vs_if/latex_tables.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":"bd78cc974c9ed25e"}},{"code_sha256_prefix":"47a59cef57459cbd","entry":"load_model","repo":"kothasuhas/understanding-forgetting","repo_kind":"official","path":"code_xnli/xnli.py","file_url":"https://github.com/kothasuhas/understanding-forgetting/blob/HEAD/code_xnli/xnli.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":"47a59cef57459cbd"}},{"code_sha256_prefix":"eaa1bb68cf7dc763","entry":"model_forward_batch","repo":"kothasuhas/understanding-forgetting","repo_kind":"official","path":"code_xnli/xnli.py","file_url":"https://github.com/kothasuhas/understanding-forgetting/blob/HEAD/code_xnli/xnli.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":"eaa1bb68cf7dc763"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}