{"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-llms-requires-more-than","title":"Position: Understanding LLMs Requires More Than Statistical Generalization","arxiv_id":"2405.01964","date":"2024-05-03","proceeding":null,"authors":["Patrik Reizinger","Szilvia Ujváry","Anna Mészáros","Anna Kerekes","Wieland Brendel","Ferenc Huszár"],"abstract":"The last decade has seen blossoming research in deep learning theory attempting to answer, \"Why does deep learning generalize?\" A powerful shift in perspective precipitated this progress: the study of overparametrized models in the interpolation regime. In this paper, we argue that another perspective shift is due, since some of the desirable qualities of LLMs are not a consequence of good statistical generalization and require a separate theoretical explanation. Our core argument relies on the observation that AR probabilistic models are inherently non-identifiable: models zero or near-zero KL divergence apart -- thus, equivalent test loss -- can exhibit markedly different behaviors. We support our position with mathematical examples and empirical observations, illustrating why non-identifiability has practical relevance through three case studies: (1) the non-identifiability of zero-shot rule extrapolation; (2) the approximate non-identifiability of in-context learning; and (3) the non-identifiability of fine-tunability. We review promising research directions focusing on LLM-relevant generalization measures, transferability, and inductive biases.","url_abs":"https://arxiv.org/abs/2405.01964v3","url_pdf":"https://arxiv.org/pdf/2405.01964v3.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-llms-requires-more-than","repo_url":"https://github.com/rpatrik96/llm-non-identifiability","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"in-context-learning","task_name":"In-Context Learning"},{"task_slug":"learning-theory","task_name":"Learning Theory"},{"task_slug":null,"task_name":"Position"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2405.01964","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.01964"}},"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/rpatrik96/llm-non-identifiability","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":4},"by_repo_kind":{"official":{"samples":4,"ran":4,"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":0,"samples":[{"code_sha256_prefix":"e98cbacccafe9522","entry":"generate_aNbN_grammar_data","repo":"rpatrik96/llm-non-identifiability","repo_kind":"official","path":"llm_non_identifiability/data.py","file_url":"https://github.com/rpatrik96/llm-non-identifiability/blob/HEAD/llm_non_identifiability/data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e98cbacccafe9522"}},{"code_sha256_prefix":"fe274e2c2938d582","entry":"generate_aNbNaN_grammar_data","repo":"rpatrik96/llm-non-identifiability","repo_kind":"official","path":"llm_non_identifiability/data.py","file_url":"https://github.com/rpatrik96/llm-non-identifiability/blob/HEAD/llm_non_identifiability/data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fe274e2c2938d582"}},{"code_sha256_prefix":"c99160bcd8350e10","entry":"generate_abN_grammar_data","repo":"rpatrik96/llm-non-identifiability","repo_kind":"official","path":"llm_non_identifiability/data.py","file_url":"https://github.com/rpatrik96/llm-non-identifiability/blob/HEAD/llm_non_identifiability/data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"c99160bcd8350e10"}},{"code_sha256_prefix":"b44c69092c71aaa3","entry":"get_tgt_mask","repo":"rpatrik96/llm-non-identifiability","repo_kind":"official","path":"llm_non_identifiability/model.py","file_url":"https://github.com/rpatrik96/llm-non-identifiability/blob/HEAD/llm_non_identifiability/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b44c69092c71aaa3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}