{"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/efficacy-of-language-model-self-play-in-non","title":"Efficacy of Language Model Self-Play in Non-Zero-Sum Games","arxiv_id":"2406.18872","date":"2024-06-27","proceeding":null,"authors":["Austen Liao","Nicholas Tomlin","Dan Klein"],"abstract":"Game-playing agents like AlphaGo have achieved superhuman performance through self-play, which is theoretically guaranteed to yield optimal policies in competitive games. However, most language tasks are partially or fully cooperative, so it is an open question whether techniques like self-play can effectively be used to improve language models. We empirically investigate this question in a negotiation game setting known as Deal or No Deal (DoND). Crucially, the objective in DoND can be modified to produce a fully cooperative game, a strictly competitive one, or anything in between. We finetune language models in self-play over multiple rounds of filtered behavior cloning in DoND for each of these objectives and evaluate them in self-play and in collaboration with humans. We find that language models improve substantially in self-play, achieving 14-17x higher scores in task reward after finetuning. Further, the trained models generalize to both cooperation and competition with humans, scoring 2.5-6x higher than base models. We view these results as an early promising sign for language model self-play in cooperative settings, despite a lack of theoretical guarantees.","url_abs":"https://arxiv.org/abs/2406.18872v2","url_pdf":"https://arxiv.org/pdf/2406.18872v2.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":"efficacy-of-language-model-self-play-in-non","repo_url":"https://github.com/nickatomlin/lm-selfplay","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[{"method_slug":"base","method_name":"BASE"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2406.18872","atlas_url":"https://app.syntology.ai/?focus=2406.18872","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2406.18872"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/nickatomlin/lm-selfplay","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":6},"by_repo_kind":{"official":{"samples":6,"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":0,"samples":[{"code_sha256_prefix":"50cf1dd8ad001a02","entry":"calculateFinalScore","repo":"nickatomlin/lm-selfplay","repo_kind":"official","path":"utils.py","file_url":"https://github.com/nickatomlin/lm-selfplay/blob/HEAD/utils.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":"50cf1dd8ad001a02"}},{"code_sha256_prefix":"d6bd8ab38d827e4c","entry":"calculateScore","repo":"nickatomlin/lm-selfplay","repo_kind":"official","path":"utils.py","file_url":"https://github.com/nickatomlin/lm-selfplay/blob/HEAD/utils.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":"d6bd8ab38d827e4c"}},{"code_sha256_prefix":"53865464ea6e4582","entry":"calculate_avg_score","repo":"nickatomlin/lm-selfplay","repo_kind":"official","path":"analysis.py","file_url":"https://github.com/nickatomlin/lm-selfplay/blob/HEAD/analysis.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":"53865464ea6e4582"}},{"code_sha256_prefix":"daa7afcdf7fc4595","entry":"check_agreement_validity","repo":"nickatomlin/lm-selfplay","repo_kind":"official","path":"utils.py","file_url":"https://github.com/nickatomlin/lm-selfplay/blob/HEAD/utils.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":"daa7afcdf7fc4595"}},{"code_sha256_prefix":"09f5d85dbba60372","entry":"parse_context","repo":"nickatomlin/lm-selfplay","repo_kind":"official","path":"play.py","file_url":"https://github.com/nickatomlin/lm-selfplay/blob/HEAD/play.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"09f5d85dbba60372"}},{"code_sha256_prefix":"521481a1b9af58fb","entry":"parse_game_context","repo":"nickatomlin/lm-selfplay","repo_kind":"official","path":"analysis.py","file_url":"https://github.com/nickatomlin/lm-selfplay/blob/HEAD/analysis.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"521481a1b9af58fb"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}