{"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/tailoring-self-rationalizers-with-multi","title":"Tailoring Self-Rationalizers with Multi-Reward Distillation","arxiv_id":"2311.02805","date":"2023-11-06","proceeding":null,"authors":["Sahana Ramnath","Brihi Joshi","Skyler Hallinan","Ximing Lu","Liunian Harold Li","Aaron Chan","Jack Hessel","Yejin Choi","Xiang Ren"],"abstract":"Large language models (LMs) are capable of generating free-text rationales to aid question answering. However, prior work 1) suggests that useful self-rationalization is emergent only at significant scales (e.g., 175B parameter GPT-3); and 2) focuses largely on downstream performance, ignoring the semantics of the rationales themselves, e.g., are they faithful, true, and helpful for humans? In this work, we enable small-scale LMs (approx. 200x smaller than GPT-3) to generate rationales that not only improve downstream task performance, but are also more plausible, consistent, and diverse, assessed both by automatic and human evaluation. Our method, MaRio (Multi-rewArd RatIOnalization), is a multi-reward conditioned self-rationalization algorithm that optimizes multiple distinct properties like plausibility, diversity and consistency. Results on five difficult question-answering datasets StrategyQA, QuaRel, OpenBookQA, NumerSense and QASC show that not only does MaRio improve task accuracy, but it also improves the self-rationalization quality of small LMs across the aforementioned axes better than a supervised fine-tuning (SFT) baseline. Extensive human evaluations confirm that MaRio rationales are preferred vs. SFT rationales, as well as qualitative improvements in plausibility and consistency.","url_abs":"https://arxiv.org/abs/2311.02805v2","url_pdf":"https://arxiv.org/pdf/2311.02805v2.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":"tailoring-self-rationalizers-with-multi","repo_url":"https://github.com/ink-usc/rationalemultirewarddistillation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"strategyqa","task_name":"StrategyQA"}],"methods":[{"method_slug":"sft","method_name":"SFT"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2311.02805","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2311.02805"}},"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/ink-usc/rationalemultirewarddistillation","reach":{"status":"ok"}}],"summary":{"ran":6,"unverified":4},"by_repo_kind":{"official":{"samples":10,"ran":6,"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":10,"samples":[{"code_sha256_prefix":"6b1269ddfa3e825b","entry":"get_accuracy","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/eval_n.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/eval_n.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":"6b1269ddfa3e825b"}},{"code_sha256_prefix":"016f3dd6b689a3e6","entry":"get_accuracy","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/eval_other_rationales.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/eval_other_rationales.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"016f3dd6b689a3e6"}},{"code_sha256_prefix":"8ab95091bfc20aec","entry":"get_control_code","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/eval_n.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/eval_n.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":"8ab95091bfc20aec"}},{"code_sha256_prefix":"6972c9dca0a0769b","entry":"get_demonstrations","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/call_gpt3.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/call_gpt3.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6972c9dca0a0769b"}},{"code_sha256_prefix":"0aa48369c05a5c8e","entry":"get_pred_labels","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/eval_n.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/eval_n.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":"0aa48369c05a5c8e"}},{"code_sha256_prefix":"98f4960217eb003e","entry":"reorganize_pools","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/clean_numersense_traindata.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/clean_numersense_traindata.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":"98f4960217eb003e"}},{"code_sha256_prefix":"b9c5ee239d0e37be","entry":"calc_and_return_flanT5_completeness_scores","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/flanT5_scores.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/flanT5_scores.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":"b9c5ee239d0e37be"}},{"code_sha256_prefix":"fbc12cd5aeca6ae7","entry":"calc_and_return_flanT5_scores","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/flanT5_scores.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/flanT5_scores.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":"fbc12cd5aeca6ae7"}},{"code_sha256_prefix":"9aaf760278c5bda9","entry":"get_rq_scores","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/rq_score.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/rq_score.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":"9aaf760278c5bda9"}},{"code_sha256_prefix":"9ceb5717399a170a","entry":"load_supervised_rq","repo":"ink-usc/rationalemultirewarddistillation","repo_kind":"official","path":"mario/rq_score.py","file_url":"https://github.com/ink-usc/rationalemultirewarddistillation/blob/HEAD/mario/rq_score.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":"9ceb5717399a170a"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}