{"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/reft-representation-finetuning-for-language","title":"ReFT: Representation Finetuning for Language Models","arxiv_id":"2404.03592","date":"2024-04-04","proceeding":null,"authors":["Zhengxuan Wu","Aryaman Arora","Zheng Wang","Atticus Geiger","Dan Jurafsky","Christopher D. Manning","Christopher Potts"],"abstract":"Parameter-efficient finetuning (PEFT) methods seek to adapt large neural models via updates to a small number of weights. However, much prior interpretability work has shown that representations encode rich semantic information, suggesting that editing representations might be a more powerful alternative. We pursue this hypothesis by developing a family of Representation Finetuning (ReFT) methods. ReFT methods operate on a frozen base model and learn task-specific interventions on hidden representations. We define a strong instance of the ReFT family, Low-rank Linear Subspace ReFT (LoReFT), and we identify an ablation of this method that trades some performance for increased efficiency. Both are drop-in replacements for existing PEFTs and learn interventions that are 15x--65x more parameter-efficient than LoRA. We showcase LoReFT on eight commonsense reasoning tasks, four arithmetic reasoning tasks, instruction-tuning, and GLUE. In all these evaluations, our ReFTs deliver the best balance of efficiency and performance, and almost always outperform state-of-the-art PEFTs. We release a generic ReFT training library publicly at https://github.com/stanfordnlp/pyreft.","url_abs":"https://arxiv.org/abs/2404.03592v3","url_pdf":"https://arxiv.org/pdf/2404.03592v3.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":"reft-representation-finetuning-for-language","repo_url":"https://github.com/stanfordnlp/pyreft","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"reft-representation-finetuning-for-language","repo_url":"https://github.com/lqtrung1998/mwp_reft","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"arithmetic-reasoning","task_name":"Arithmetic Reasoning"}],"methods":[{"method_slug":"base","method_name":"BASE"},{"method_slug":null,"method_name":"Library"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2404.03592","atlas_url":"https://app.syntology.ai/?focus=2404.03592","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2404.03592"},"calls":[{"tool":"get_harvested_code_for_paper","arguments":{"arxiv_id":"2404.03592"},"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":"2404.03592"},"call":"get_harvested_code_for_paper(arxiv_id=\"2404.03592\")"},{"tool":"get_code","arguments_in":{"code_sha256":"syntology.samples[].code_sha256"},"call":"get_code(code_sha256=\"…\")"},{"tool":"get_citation_path","arguments":{"paper_1":"2404.03592"},"arguments_in":{"paper_2":"another paper's arXiv id or title"},"call":"get_citation_path(paper_1=\"2404.03592\", paper_2=\"…\")"},{"tool":"get_concepts_for_paper","arguments":{"arxiv_id":"2404.03592"},"call":"get_concepts_for_paper(arxiv_id=\"2404.03592\")"}],"by_repo":[{"repo":"stanfordnlp/pyreft","kind":"official","samples":1,"ran":1,"instrument":0,"unverified":0},{"repo":"lqtrung1998/mwp_reft","kind":"listed","samples":13,"ran":5,"instrument":0,"unverified":8}],"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/2404.03592.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/2404.03592.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/stanfordnlp/pyreft","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/lqtrung1998/mwp_reft","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":3,"ran_honours":2,"ran_violates":1,"unverified":8},"n_samples":14,"n_ran":6,"n_constructed":0,"n_ran_checked":6,"n_instrument":0,"n_unverified":8,"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":1,"ran":1,"n_ran_checked":1,"n_instrument":0,"repositories":1},"listed":{"samples":13,"ran":5,"n_ran_checked":5,"n_instrument":0,"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":12,"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":12,"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":6,"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":"2bb5bd3526f55ecb","code_sha256_prefix":"2bb5bd3526f55ecb","entry":"collate_fn","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"sampling.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/sampling.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"features:list/1/dict, tokenizer:MockTokenizer/0/?","fixture_out_type":"dict","level_recorded_at":"2026-09-25T16:13:10.526492+00:00","code_page":"/code/collate-fn","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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"2bb5bd3526f55ecb"}},{"code_sha256":"1cd5252a6c21c50e","code_sha256_prefix":"1cd5252a6c21c50e","entry":"count_parameters","repo":"stanfordnlp/pyreft","repo_kind":"official","path":"pyreft/reft_model.py","file_url":"https://github.com/stanfordnlp/pyreft/blob/HEAD/pyreft/reft_model.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"signature":"model:callable/0/?","fixture_out_type":"int","level_recorded_at":"2026-09-22T14:31:46.068493+00:00","code_page":"/code/count-parameters","census":null,"licence":"Apache-2.0","inline_ok":true,"inline_ok_link":true,"withheld_clause":null,"mcp_get_code":{"code_sha256":"1cd5252a6c21c50e"}},{"code_sha256":"f21a5f0d5bae0c14","code_sha256_prefix":"f21a5f0d5bae0c14","entry":"floatify","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"src/utils.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/src/utils.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"s:str/0/str","fixture_out_type":"float","level_recorded_at":"2026-09-25T23:22:42.264232+00:00","retrieve_with":{"tool":"get_code","code_sha256":"f21a5f0d5bae0c14"},"code_page":"/code/floatify","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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"f21a5f0d5bae0c14"}},{"code_sha256":"c3b86a3bceec86cd","code_sha256_prefix":"c3b86a3bceec86cd","entry":"is_numeric","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"src/utils.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/src/utils.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"value:int/0/int","fixture_out_type":"bool","level_recorded_at":"2026-09-25T16:13:10.526492+00:00","retrieve_with":{"tool":"get_code","code_sha256":"c3b86a3bceec86cd"},"code_page":"/code/is-numeric","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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"c3b86a3bceec86cd"}},{"code_sha256":"8b9f60df8d2c8a82","code_sha256_prefix":"8b9f60df8d2c8a82","entry":"process_code","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"src/python_engine.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/src/python_engine.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"signature":"code_gen:str/0/str, truncate_first_return:bool/0/bool","fixture_out_type":"str","level_recorded_at":"2026-09-25T16:13:10.526492+00:00","retrieve_with":{"tool":"get_code","code_sha256":"8b9f60df8d2c8a82"},"code_page":"/code/process-code","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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"8b9f60df8d2c8a82"}},{"code_sha256":"951fe709737c50b9","code_sha256_prefix":"951fe709737c50b9","entry":"tokenize_fn","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"sampling.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/sampling.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"signature":"examples:dict/0/?, tokenizer:MinimalTokenizer/0/?, max_length:int/0/int, src_name:str/0/str, engine:str/0/str","fixture_out_type":"dict","level_recorded_at":"2026-09-25T16:13:10.526492+00:00","code_page":"/code/tokenize-fn","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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"951fe709737c50b9"}},{"code_sha256":"32851be78f5ecd4c","code_sha256_prefix":"32851be78f5ecd4c","entry":"do_gather","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"src/utils.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/src/utils.py","link_basis":"harvester_set","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":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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"32851be78f5ecd4c"}},{"code_sha256":"7fcae9f5e120bafe","code_sha256_prefix":"7fcae9f5e120bafe","entry":"eval_code","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"src/python_engine.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/src/python_engine.py","link_basis":"harvester_set","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":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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"7fcae9f5e120bafe"}},{"code_sha256":"c8164df09b7d7a68","code_sha256_prefix":"c8164df09b7d7a68","entry":"evaluate_rerank","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"train_reward_model.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/train_reward_model.py","link_basis":"harvester_set","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":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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"c8164df09b7d7a68"}},{"code_sha256":"4d434f6e5fd591b5","code_sha256_prefix":"4d434f6e5fd591b5","entry":"prepare_datasets_and_data_loaders","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"train_reward_model.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/train_reward_model.py","link_basis":"harvester_set","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":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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"4d434f6e5fd591b5"}},{"code_sha256":"9f0489fe01df9539","code_sha256_prefix":"9f0489fe01df9539","entry":"prepare_datasets_and_data_loaders","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"train_sl_model.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/train_sl_model.py","link_basis":"harvester_set","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":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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"9f0489fe01df9539"}},{"code_sha256":"449e6a4081f8d019","code_sha256_prefix":"449e6a4081f8d019","entry":"prepare_datasets_and_data_loaders","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"train_rl_reft.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/train_rl_reft.py","link_basis":"harvester_set","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":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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"449e6a4081f8d019"}},{"code_sha256":"d0eff44d0f627c24","code_sha256_prefix":"d0eff44d0f627c24","entry":"prepare_datasets_and_data_loaders","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"train_sft_model.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/train_sft_model.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":"d0eff44d0f627c24"}},{"code_sha256":"d3b8681876990bfa","code_sha256_prefix":"d3b8681876990bfa","entry":"train_one_epoch","repo":"lqtrung1998/mwp_reft","repo_kind":"listed","path":"train_reward_model.py","file_url":"https://github.com/lqtrung1998/mwp_reft/blob/HEAD/train_reward_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"code_page":"/code/train-one-epoch","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 WNQzhu/CARFT, and GitHub detected no licence in that repository when it was checked","mcp_get_code":{"code_sha256":"d3b8681876990bfa"}}]},"code_links_note":null,"arxiv_metadata":null,"syntology_extracted_results":null,"record_sha256":"e794561c0d34894d8250720efa70215f5feb2514848e514e5337aa725bc39e6c","record_changed_at":"2026-10-04","record_changed_at_basis":"changed"}