{"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/comal-a-convergent-meta-algorithm-for","title":"COMAL: A Convergent Meta-Algorithm for Aligning LLMs with General Preferences","arxiv_id":"2410.23223","date":"2024-10-30","proceeding":null,"authors":["Yixin Liu","Argyris Oikonomou","Weiqiang Zheng","Yang Cai","Arman Cohan"],"abstract":"Many alignment methods, including reinforcement learning from human feedback (RLHF), rely on the Bradley-Terry reward assumption, which is insufficient to capture the full range of general human preferences. To achieve robust alignment with general preferences, we model the alignment problem as a two-player zero-sum game, where the Nash equilibrium policy guarantees a 50% win rate against any competing policy. However, previous algorithms for finding the Nash policy either diverge or converge to a Nash policy in a modified game, even in a simple synthetic setting, thereby failing to maintain the 50% win rate guarantee against all other policies. We propose a meta-algorithm, Convergent Meta Alignment Algorithm (COMAL), for language model alignment with general preferences, inspired by convergent algorithms in game theory. Theoretically, we prove that our meta-algorithm converges to an exact Nash policy in the last iterate. Additionally, our meta-algorithm is simple and can be integrated with many existing methods designed for RLHF and preference optimization with minimal changes. Experimental results demonstrate the effectiveness of the proposed framework when combined with existing preference policy optimization methods.","url_abs":"https://arxiv.org/abs/2410.23223v1","url_pdf":"https://arxiv.org/pdf/2410.23223v1.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":"comal-a-convergent-meta-algorithm-for","repo_url":"https://github.com/yale-nlp/comal","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2410.23223","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.23223"}},"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/yale-nlp/comal","reach":{"status":"ok"}}],"summary":{"unverified":10},"by_repo_kind":{"official":{"samples":10,"ran":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":10,"samples":[{"code_sha256_prefix":"dee796ba8ebdf795","entry":"apply_chat_template_qwen","repo":"yale-nlp/comal","repo_kind":"official","path":"data_processing.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/data_processing.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":"dee796ba8ebdf795"}},{"code_sha256_prefix":"171a6bd18e3c8aab","entry":"collate_nash_preference_qwen","repo":"yale-nlp/comal","repo_kind":"official","path":"data_utils.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/data_utils.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":"171a6bd18e3c8aab"}},{"code_sha256_prefix":"bf2534585f308a5a","entry":"collate_preference_base_qwen","repo":"yale-nlp/comal","repo_kind":"official","path":"data_utils.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/data_utils.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":"bf2534585f308a5a"}},{"code_sha256_prefix":"4eb403373472e890","entry":"collate_preference_qwen","repo":"yale-nlp/comal","repo_kind":"official","path":"data_utils.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/data_utils.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":"4eb403373472e890"}},{"code_sha256_prefix":"a71ddf2f2e1209ca","entry":"dpo_loss","repo":"yale-nlp/comal","repo_kind":"official","path":"losses.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/losses.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":"a71ddf2f2e1209ca"}},{"code_sha256_prefix":"7135ae940c5f72ac","entry":"ipo_loss","repo":"yale-nlp/comal","repo_kind":"official","path":"losses.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/losses.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":"7135ae940c5f72ac"}},{"code_sha256_prefix":"a2b71f28d73fefc6","entry":"is_port_in_use","repo":"yale-nlp/comal","repo_kind":"official","path":"utils.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/utils.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":"a2b71f28d73fefc6"}},{"code_sha256_prefix":"ee92b1bdda006b64","entry":"nash_loss","repo":"yale-nlp/comal","repo_kind":"official","path":"losses.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/losses.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":"ee92b1bdda006b64"}},{"code_sha256_prefix":"1ab6ea4de9ffa9aa","entry":"read_json","repo":"yale-nlp/comal","repo_kind":"official","path":"utils.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/utils.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":"1ab6ea4de9ffa9aa"}},{"code_sha256_prefix":"e8f03c37313f8363","entry":"test","repo":"yale-nlp/comal","repo_kind":"official","path":"mle.py","file_url":"https://github.com/yale-nlp/comal/blob/HEAD/mle.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":"e8f03c37313f8363"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}