{"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/statistical-impossibility-and-possibility-of","title":"Statistical Impossibility and Possibility of Aligning LLMs with Human Preferences: From Condorcet Paradox to Nash Equilibrium","arxiv_id":"2503.10990","date":"2025-03-14","proceeding":null,"authors":["Kaizhao Liu","Qi Long","Zhekun Shi","Weijie J. Su","Jiancong Xiao"],"abstract":"Aligning large language models (LLMs) with diverse human preferences is critical for ensuring fairness and informed outcomes when deploying these models for decision-making. In this paper, we seek to uncover fundamental statistical limits concerning aligning LLMs with human preferences, with a focus on the probabilistic representation of human preferences and the preservation of diverse preferences in aligned LLMs. We first show that human preferences can be represented by a reward model if and only if the preference among LLM-generated responses is free of any Condorcet cycle. Moreover, we prove that Condorcet cycles exist with probability converging to one exponentially fast under a probabilistic preference model, thereby demonstrating the impossibility of fully aligning human preferences using reward-based approaches such as reinforcement learning from human feedback. Next, we explore the conditions under which LLMs would employ mixed strategies -- meaning they do not collapse to a single response -- when aligned in the limit using a non-reward-based approach, such as Nash learning from human feedback (NLHF). We identify a necessary and sufficient condition for mixed strategies: the absence of a response that is preferred over all others by a majority. As a blessing, we prove that this condition holds with high probability under the probabilistic preference model, thereby highlighting the statistical possibility of preserving minority preferences without explicit regularization in aligning LLMs. Finally, we leverage insights from our statistical results to design a novel, computationally efficient algorithm for finding Nash equilibria in aligning LLMs with NLHF. Our experiments show that Llama-3.2-1B, aligned with our algorithm, achieves a win rate of 60.55\\% against the base model.","url_abs":"https://arxiv.org/abs/2503.10990v1","url_pdf":"https://arxiv.org/pdf/2503.10990v1.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":"statistical-impossibility-and-possibility-of","repo_url":"https://github.com/szk123456789/Nash_RS","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"fairness","task_name":"Fairness"}],"methods":[{"method_slug":"base","method_name":"BASE"},{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2503.10990","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2503.10990"}},"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/szk123456789/Nash_RS","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":3,"ran_draft_wrong":1,"unverified":4},"by_repo_kind":{"official":{"samples":8,"ran":4,"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":"60d73ddf6afc1bc3","entry":"compute_approx_kl","repo":"szk123456789/Nash_RS","repo_kind":"official","path":"openrlhf/models/utils.py","file_url":"https://github.com/szk123456789/Nash_RS/blob/HEAD/openrlhf/models/utils.py","link_basis":"harvester_set","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":"60d73ddf6afc1bc3"}},{"code_sha256_prefix":"a9fb683d7652195d","entry":"get_seqlens_in_batch","repo":"szk123456789/Nash_RS","repo_kind":"official","path":"openrlhf/models/packing_utils.py","file_url":"https://github.com/szk123456789/Nash_RS/blob/HEAD/openrlhf/models/packing_utils.py","link_basis":"harvester_set","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":"a9fb683d7652195d"}},{"code_sha256_prefix":"08e261d2bbc45dbc","entry":"get_unpad_data","repo":"szk123456789/Nash_RS","repo_kind":"official","path":"openrlhf/models/packing_utils.py","file_url":"https://github.com/szk123456789/Nash_RS/blob/HEAD/openrlhf/models/packing_utils.py","link_basis":"harvester_set","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":"08e261d2bbc45dbc"}},{"code_sha256_prefix":"959df7824bf09cc1","entry":"log_probs_from_logits","repo":"szk123456789/Nash_RS","repo_kind":"official","path":"openrlhf/models/utils.py","file_url":"https://github.com/szk123456789/Nash_RS/blob/HEAD/openrlhf/models/utils.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"959df7824bf09cc1"}},{"code_sha256_prefix":"0d3d2cb3519d44c9","entry":"CalculatePreference","repo":"szk123456789/Nash_RS","repo_kind":"official","path":"openrlhf/models/NashRS_utils.py","file_url":"https://github.com/szk123456789/Nash_RS/blob/HEAD/openrlhf/models/NashRS_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0d3d2cb3519d44c9"}},{"code_sha256_prefix":"477b4c962ccae21e","entry":"compute_approx_kl","repo":"szk123456789/Nash_RS","repo_kind":"official","path":"openrlhf/models/NashRS_utils.py","file_url":"https://github.com/szk123456789/Nash_RS/blob/HEAD/openrlhf/models/NashRS_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"477b4c962ccae21e"}},{"code_sha256_prefix":"99a2d91ea5aed978","entry":"compute_reward","repo":"szk123456789/Nash_RS","repo_kind":"official","path":"openrlhf/models/utils.py","file_url":"https://github.com/szk123456789/Nash_RS/blob/HEAD/openrlhf/models/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"99a2d91ea5aed978"}},{"code_sha256_prefix":"e93117d99aa6b210","entry":"reset_ring_attn_position_ids","repo":"szk123456789/Nash_RS","repo_kind":"official","path":"openrlhf/models/ring_attn_utils.py","file_url":"https://github.com/szk123456789/Nash_RS/blob/HEAD/openrlhf/models/ring_attn_utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e93117d99aa6b210"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}