{"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/simpo-simple-preference-optimization-with-a","title":"SimPO: Simple Preference Optimization with a Reference-Free Reward","arxiv_id":"2405.14734","date":"2024-05-23","proceeding":null,"authors":["Yu Meng","Mengzhou Xia","Danqi Chen"],"abstract":"Direct Preference Optimization (DPO) is a widely used offline preference optimization algorithm that reparameterizes reward functions in reinforcement learning from human feedback (RLHF) to enhance simplicity and training stability. In this work, we propose SimPO, a simpler yet more effective approach. The effectiveness of SimPO is attributed to a key design: using the average log probability of a sequence as the implicit reward. This reward formulation better aligns with model generation and eliminates the need for a reference model, making it more compute and memory efficient. Additionally, we introduce a target reward margin to the Bradley-Terry objective to encourage a larger margin between the winning and losing responses, further improving the algorithm's performance. We compare SimPO to DPO and its latest variants across various state-of-the-art training setups, including both base and instruction-tuned models such as Mistral, Llama 3, and Gemma 2. We evaluate on extensive chat-based evaluation benchmarks, including AlpacaEval 2, MT-Bench, and Arena-Hard. Our results demonstrate that SimPO consistently and significantly outperforms existing approaches without substantially increasing response length. Specifically, SimPO outperforms DPO by up to 6.4 points on AlpacaEval 2 and by up to 7.5 points on Arena-Hard. Our top-performing model, built on Gemma-2-9B-it, achieves a 72.4% length-controlled win rate on AlpacaEval 2, a 59.1% win rate on Arena-Hard, and ranks 1st on Chatbot Arena among <10B models with real user votes.","url_abs":"https://arxiv.org/abs/2405.14734v3","url_pdf":"https://arxiv.org/pdf/2405.14734v3.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":"simpo-simple-preference-optimization-with-a","repo_url":"https://github.com/princeton-nlp/simpo","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"simpo-simple-preference-optimization-with-a","repo_url":"https://github.com/openpsi-project/realhf","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"chatbot","task_name":"Chatbot"},{"task_slug":"instruction-following","task_name":"Instruction Following"}],"methods":[{"method_slug":"base","method_name":"BASE"},{"method_slug":"dpo","method_name":"DPO"},{"method_slug":"llama","method_name":"LLaMA"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2405.14734","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.14734"}},"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":"deterministic:regex_extraction","url":"https://github.com/princeton-nlp/SimPO","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/openpsi-project/realhf","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/princeton-nlp/simpo","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":4,"ran_draft_wrong":1,"ran_violates":1},"by_repo_kind":{"official":{"samples":5,"ran":5,"repositories":1},"listed":{"samples":1,"ran":1,"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":"9be97bcdb51fd634","entry":"apply_chat_template","repo":"princeton-nlp/SimPO","repo_kind":"official","path":"alignment/data.py","file_url":"https://github.com/princeton-nlp/SimPO/blob/HEAD/alignment/data.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":"9be97bcdb51fd634"}},{"code_sha256_prefix":"6b05e6f1ad3cb9b6","entry":"dpo_loss","repo":"openpsi-project/realhf","repo_kind":"listed","path":"realhf/impl/model/utils/dpo_functional.py","file_url":"https://github.com/openpsi-project/realhf/blob/HEAD/realhf/impl/model/utils/dpo_functional.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"6b05e6f1ad3cb9b6"}},{"code_sha256_prefix":"0edeacd7a9034f53","entry":"extract_docstring","repo":"princeton-nlp/SimPO","repo_kind":"official","path":"alignment/decontaminate.py","file_url":"https://github.com/princeton-nlp/SimPO/blob/HEAD/alignment/decontaminate.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":"0edeacd7a9034f53"}},{"code_sha256_prefix":"4253aac010c4504f","entry":"is_openai_format","repo":"princeton-nlp/SimPO","repo_kind":"official","path":"alignment/data.py","file_url":"https://github.com/princeton-nlp/SimPO/blob/HEAD/alignment/data.py","link_basis":"harvester_set","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4253aac010c4504f"}},{"code_sha256_prefix":"a3791c5c2185e013","entry":"load_dataset_column","repo":"princeton-nlp/SimPO","repo_kind":"official","path":"alignment/decontaminate.py","file_url":"https://github.com/princeton-nlp/SimPO/blob/HEAD/alignment/decontaminate.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":"a3791c5c2185e013"}},{"code_sha256_prefix":"f43e59d239019e86","entry":"normalize_whitespace","repo":"princeton-nlp/SimPO","repo_kind":"official","path":"alignment/decontaminate.py","file_url":"https://github.com/princeton-nlp/SimPO/blob/HEAD/alignment/decontaminate.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":"f43e59d239019e86"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}