{"about":{"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.","site":"https://codewithpapers.app","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","syntology":{"site":"https://syntology.ai","developers":"https://syntology.ai/developers","mcp":{"server":"https://syntology.ai/mcp","transport":"streamable-http","server_card":"https://syntology.ai/.well-known/mcp/server-card.json","auth":{"type":"trial token, no account","trial_token":"https://syntology.ai/api/oauth/trial/token","method":"POST","docs":"https://syntology.ai/developers"}},"have":"https://syntology.ai/api/graph/have?x=<method, arXiv id or title> (free, answers coverage only)","paper_base":"https://syntology.ai/paper/","atlas_base":"https://app.syntology.ai/?focus="},"machine_readable":[{"url":"https://codewithpapers.app/llms.txt","what":"the machine catalog: every machine-readable file, counted"},{"url":"https://codewithpapers.app/index/manifest.json","what":"paper-to-code index by arXiv id, with Syntology's counts"},{"url":"https://codewithpapers.app/search/manifest.json","what":"site search index (titles, authors) and its files"},{"url":"https://codewithpapers.app/download","what":"bulk files: Syntology's layer, described there"},{"url":"https://codewithpapers.app/build_manifest.json","what":"the build record: inputs, counts, exclusions, probes"}]},"url":"/paper/learning-dynamics-of-llm-finetuning","title":"Learning Dynamics of LLM Finetuning","arxiv_id":"2407.10490","date":"2024-07-15","proceeding":null,"authors":["Yi Ren","Danica J. Sutherland"],"abstract":"Learning dynamics, which describes how the learning of specific training examples influences the model's predictions on other examples, gives us a powerful tool for understanding the behavior of deep learning systems. We study the learning dynamics of large language models during different types of finetuning, by analyzing the step-wise decomposition of how influence accumulates among different potential responses. Our framework allows a uniform interpretation of many interesting observations about the training of popular algorithms for both instruction tuning and preference tuning. In particular, we propose a hypothetical explanation of why specific types of hallucination are strengthened after finetuning, e.g., the model might use phrases or facts in the response for question B to answer question A, or the model might keep repeating similar simple phrases when generating responses. We also extend our framework and highlight a unique \"squeezing effect\" to explain a previously observed phenomenon in off-policy direct preference optimization (DPO), where running DPO for too long makes even the desired outputs less likely. This framework also provides insights into where the benefits of on-policy DPO and other variants come from. The analysis not only provides a novel perspective of understanding LLM's finetuning but also inspires a simple, effective method to improve alignment performance.","url_abs":"https://arxiv.org/abs/2407.10490v2","url_pdf":"https://arxiv.org/pdf/2407.10490v2.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":"learning-dynamics-of-llm-finetuning","repo_url":"https://github.com/joshua-ren/learning_dynamics_llm","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"hallucination","task_name":"Hallucination"}],"methods":[{"method_slug":"dpo","method_name":"DPO"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2407.10490","atlas_url":"https://app.syntology.ai/?focus=2407.10490","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2407.10490"},"calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2407.10490"},"returns":"the samples harvested from this paper, each with its run record; it does not verify the paper"},{"tool":"get_code","arguments_in":"samples[].code_sha256","returns":"one stored sample by its code_sha256; not shown on this site"}]},"developers":"https://syntology.ai/developers","held":true,"reason":"in_graph_layer","checked_against":null,"agent_calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2407.10490"},"call":"get_harvested_code_for_paper(arxiv_id=\"2407.10490\")"},{"tool":"get_code","arguments_in":{"code_sha256":"syntology.samples[].code_sha256"},"call":"get_code(code_sha256=\"…\")"},{"tool":"get_citation_path","arguments":{"paper_1":"2407.10490"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2407.10490\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2407.10490"},"call":"get_concepts_for_paper(arxiv_id=\"2407.10490\")"}],"by_repo":[{"repo":"joshua-ren/learning_dynamics_llm","kind":"official","samples":10,"ran":7,"instrument":5,"unverified":3}],"by_repo_is":"one row per repository in the archive's code links, Syntology's graph links or the samples; samples 0 means no sample is linked to this paper, a statement about Syntology's coverage, not about the repository: with harvested_for_other_papers present, Syntology harvested the repository for that many other papers, and without it, Syntology harvested nothing from it; placed_by_identical_code counts samples whose link records no repository, placed here because identical code was harvested from this repository for another paper; ran counts executions on a synthesized input and instrument counts failures of Syntology's instrument, not of the code","read_at":"2026-10-05T00:36:11+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","ran_record":{"url":"https://syntology.ai/api/ran/2407.10490.json","schema":"syntology.ran_record/1","counts_crosswalk":{"n_samples":"syntology.counts.lifted","n_ran":"syntology.counts.executed","n_ran_checked":"syntology.counts.checked","n_instrument":"syntology.counts.instrument_failures","n_unverified":"syntology.counts.no_recorded_run"},"note":"live record, measured at request time; may be newer than this page's graph read","badge":"https://syntology.ai/api/ran/2407.10490.svg"},"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/joshua-ren/learning_dynamics_llm","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_draft_wrong":4,"ran":2,"ran_fixture":1,"unverified":3},"n_samples":10,"n_ran":7,"n_constructed":0,"n_ran_checked":2,"n_instrument":5,"n_unverified":3,"n_instrument_is":"failures of Syntology's instrument, not of the code","every_run_is_an_instrument_failure":false,"key_notes":{"n_ran_checked":"legacy name, kept unchanged so existing readers do not break: it counts the samples that ran with no instrument failure (honoured, violated, and ran with no contract checked); it does not mean a contract was checked, and the pages print it as 'K with no instrument failure', not 'K checked'","n_constructed":"a sub-count of the samples that ran, never subtracted from them and never a failure: an executed sample whose run returned an instance of its own class (fixture_out_type equals the entry name): the run built an object and did not compute a result (Syntology's RAN record, counts.constructed)"},"by_repo_kind":{"official":{"samples":10,"ran":7,"n_ran_checked":2,"n_instrument":5,"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":2,"inline_ok_is":"inline_ok is true only when both licence gates allow Syntology to serve this code's text: this paper's link (inline_ok_link, from the licence of this paper's copy) and the stored copy of the same code that Syntology serves, which may have been first harvested from another repository. withheld_clause is Syntology's own wording of which gate keeps the text a pointer and why, and is null when inline_ok is true. This site shows no code text either way; n_pointer_only_for_licence counts samples with inline_ok false (both gates), n_pointer_only_link_allows those of them whose inline_ok_link is true","n_pointer_only_link_allows":2,"fixture_fields_is":{"signature_and_fixture_out_type":"the arguments Syntology's fixture called the sample with, each as name:type/rank/element kind (? where none was recorded), and the type of what the fixture's run returned, exactly as the harness recorded them; the arguments and the returned type were recorded by two separate runs of the same fixture. No output value is shown, and a returned type says nothing about whether the output was right","level_recorded_at":"when Syntology's graph wrote the sample's verification level, not when the code executed; absent where the level was written before the date was kept, and never filled in with the build date","present_on":"samples that ran only; a key is absent, never empty, where the graph records nothing"},"n_ran_with_level_date":7,"n_ran_without_level_date":0,"n_ran_signature_not_shown":0,"levels_written_between":{"first":"2026-09-22","last":"2026-09-25","is":"the first and last UTC dates on which Syntology's graph wrote a verification level for a sample on this page that ran, not when the code executed"},"code_page_min_papers":5,"code_page_is":"a page for the entry name across every paper's harvested code; the name is a string, not an identity","samples":[{"code_sha256":"12f6803880f9b17e","code_sha256_prefix":"12f6803880f9b17e","entry":"extract_anthropic_prompt","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/preference_datasets.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/preference_datasets.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"signature":"prompt_and_response:str/0/str","fixture_out_type":"str","level_recorded_at":"2026-09-22T14:31:46.068493+00:00","code_page":"/code/extract-anthropic-prompt","census":null,"licence":"Apache-2.0","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from junkangwu/beta-DPO, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"12f6803880f9b17e"}},{"code_sha256":"950dcf3ad144388a","code_sha256_prefix":"950dcf3ad144388a","entry":"get_api_base_url","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/gen_dataset.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/gen_dataset.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"platform:str/0/str","fixture_out_type":"str","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"950dcf3ad144388a"}},{"code_sha256":"7d82ad82e19dab04","code_sha256_prefix":"7d82ad82e19dab04","entry":"get_hh","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/preference_datasets.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/preference_datasets.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"split:str/0/str, silent:bool/0/bool, data_dir:str/0/str","fixture_out_type":"list","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"7d82ad82e19dab04"}},{"code_sha256":"8ab43db002bf984d","code_sha256_prefix":"8ab43db002bf984d","entry":"get_local_dir","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/utils.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/utils.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"prefixes_to_resolve:list/1/str","fixture_out_type":"str","level_recorded_at":"2026-09-25T15:11:32.249708+00:00","code_page":"/code/get-local-dir","census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"8ab43db002bf984d"}},{"code_sha256":"10ab71f27ffd7de8","code_sha256_prefix":"10ab71f27ffd7de8","entry":"get_local_run_dir","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/utils.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/utils.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"exp_name:str/0/str, local_dirs:list/1/str","fixture_out_type":"str","level_recorded_at":"2026-09-24T15:16:18.137819+00:00","code_page":"/code/get-local-run-dir","census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"10ab71f27ffd7de8"}},{"code_sha256":"e7ce4c19888c6df6","code_sha256_prefix":"e7ce4c19888c6df6","entry":"get_remote_file","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/utils.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/utils.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"remote_path:str/0/str, local_path:str/0/str","fixture_out_type":"str","level_recorded_at":"2026-09-25T15:11:32.249708+00:00","code_page":"/code/get-remote-file","census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"e7ce4c19888c6df6"}},{"code_sha256":"c79211fc0225a27f","code_sha256_prefix":"c79211fc0225a27f","entry":"preference_loss","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/trainers.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/trainers.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"policy_chosen_logps:Tensor/1/float, policy_rejected_logps:Tensor/1/float, reference_chosen_logps:Tensor/1/float, reference_rejected_logps:Tensor/1/float, beta:float/0/float, label_smoothing:float/0/float, ipo:bool/0/bool, reference_free:bool/0/bool","fixture_out_type":"tuple","level_recorded_at":"2026-09-22T14:31:46.068493+00:00","retrieve_with":{"tool":"get_code","code_sha256":"c79211fc0225a27f"},"code_page":"/code/preference-loss","census":null,"licence":"Apache-2.0","inline_ok":false,"inline_ok_link":true,"withheld_clause":"this paper's link allows the body inline (Apache-2.0), but get_code serves the stored copy, first harvested from icip-cas/Word-Deed-Consistency-Test, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"c79211fc0225a27f"}},{"code_sha256":"8fa0d07c713b5722","code_sha256_prefix":"8fa0d07c713b5722","entry":"get_dataset","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/preference_datasets.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/preference_datasets.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":"/code/get-dataset","census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"8fa0d07c713b5722"}},{"code_sha256":"d6be0aae15f087e3","code_sha256_prefix":"d6be0aae15f087e3","entry":"get_gpt_completion","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/gen_dataset.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/gen_dataset.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"d6be0aae15f087e3"}},{"code_sha256":"a5e707e0c89d915f","code_sha256_prefix":"a5e707e0c89d915f","entry":"get_original_helpful","repo":"joshua-ren/learning_dynamics_llm","repo_kind":"official","path":"src/gen_dataset.py","file_url":"https://github.com/joshua-ren/learning_dynamics_llm/blob/HEAD/src/gen_dataset.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":null,"census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"a5e707e0c89d915f"}}]},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"a211193ed2169748c72a457ec4b2d7eab8fd2d46a82045d5edb3f9bf4cd4bb98","record_changed_at":"2026-10-04","record_changed_at_basis":"changed"}