{"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/reward-augmented-data-enhances-direct","title":"Reward-Augmented Data Enhances Direct Preference Alignment of LLMs","arxiv_id":"2410.08067","date":"2024-10-10","proceeding":null,"authors":["Shenao Zhang","Zhihan Liu","Boyi Liu","Yufeng Zhang","Yingxiang Yang","Yongfei Liu","Liyu Chen","Tao Sun","Zhaoran Wang"],"abstract":"Preference alignment in Large Language Models (LLMs) has significantly improved their ability to adhere to human instructions and intentions. However, existing direct alignment algorithms primarily focus on relative preferences and often overlook the qualitative aspects of responses, despite having access to preference data that includes reward scores from judge models during AI feedback. Striving to maximize the implicit reward gap between the chosen and the slightly inferior rejected responses can cause overfitting and unnecessary unlearning of the high-quality rejected responses. The unawareness of the reward scores also drives the LLM to indiscriminately favor the low-quality chosen responses and fail to generalize to optimal responses that are sparse in data. To overcome these shortcomings, our study introduces reward-conditioned LLM policies that discern and learn from the entire spectrum of response quality within the dataset, helping extrapolate to more optimal regions. We propose an effective yet simple data relabeling method that conditions the preference pairs on quality scores to construct a reward-augmented dataset. The experiments across various benchmarks and diverse models demonstrate that our approach consistently boosts DPO by a considerable margin. Through comprehensive ablation studies, we demonstrate that our method not only maximizes the utility of preference data but also mitigates the issue of unlearning, demonstrating its broad effectiveness beyond mere data expansion. Our code is available at https://github.com/shenao-zhang/reward-augmented-preference.","url_abs":"https://arxiv.org/abs/2410.08067v6","url_pdf":"https://arxiv.org/pdf/2410.08067v6.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":"reward-augmented-data-enhances-direct","repo_url":"https://github.com/shenao-zhang/reward-augmented-preference","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"instruction-following","task_name":"Instruction Following"}],"methods":[{"method_slug":"dpo","method_name":"DPO"},{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2410.08067","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.08067"}},"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/shenao-zhang/reward-augmented-preference","reach":null}],"summary":{"ran_draft_wrong":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":"34ad32ae863a3812","entry":"add_neg_goal","repo":"shenao-zhang/reward-augmented-preference","repo_kind":"official","path":"scripts/reward_augmentation.py","file_url":"https://github.com/shenao-zhang/reward-augmented-preference/blob/HEAD/scripts/reward_augmentation.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"34ad32ae863a3812"}},{"code_sha256_prefix":"0674aa408dbedf57","entry":"add_pos_goal","repo":"shenao-zhang/reward-augmented-preference","repo_kind":"official","path":"scripts/reward_augmentation.py","file_url":"https://github.com/shenao-zhang/reward-augmented-preference/blob/HEAD/scripts/reward_augmentation.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"0674aa408dbedf57"}},{"code_sha256_prefix":"1290bdbfd3cfa6e5","entry":"format_prompt","repo":"shenao-zhang/reward-augmented-preference","repo_kind":"official","path":"scripts/preprocess.py","file_url":"https://github.com/shenao-zhang/reward-augmented-preference/blob/HEAD/scripts/preprocess.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"1290bdbfd3cfa6e5"}},{"code_sha256_prefix":"2f76390e5e689883","entry":"get_pairwise_completions","repo":"shenao-zhang/reward-augmented-preference","repo_kind":"official","path":"scripts/preprocess.py","file_url":"https://github.com/shenao-zhang/reward-augmented-preference/blob/HEAD/scripts/preprocess.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2f76390e5e689883"}},{"code_sha256_prefix":"c5e24c10f8edc3f7","entry":"remove_last_step_for_rl","repo":"shenao-zhang/reward-augmented-preference","repo_kind":"official","path":"scripts/preprocess.py","file_url":"https://github.com/shenao-zhang/reward-augmented-preference/blob/HEAD/scripts/preprocess.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c5e24c10f8edc3f7"}},{"code_sha256_prefix":"636da2fc25ece4cf","entry":"system_prompt","repo":"shenao-zhang/reward-augmented-preference","repo_kind":"official","path":"scripts/reward_augmentation.py","file_url":"https://github.com/shenao-zhang/reward-augmented-preference/blob/HEAD/scripts/reward_augmentation.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"636da2fc25ece4cf"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}