{"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/rovrm-a-robust-visual-reward-model-optimized","title":"RoVRM: A Robust Visual Reward Model Optimized via Auxiliary Textual Preference Data","arxiv_id":"2408.12109","date":"2024-08-22","proceeding":null,"authors":["Chenglong Wang","Yang Gan","Yifu Huo","Yongyu Mu","Murun Yang","Qiaozhi He","Tong Xiao","Chunliang Zhang","Tongran Liu","Quan Du","Di Yang","Jingbo Zhu"],"abstract":"Large vision-language models (LVLMs) often fail to align with human preferences, leading to issues like generating misleading content without proper visual context (also known as hallucination). A promising solution to this problem is using human-preference alignment techniques, such as best-of-n sampling and reinforcement learning. However, these techniques face the difficulty arising from the scarcity of visual preference data, which is required to train a visual reward model (VRM). In this work, we continue the line of research. We present a Robust Visual Reward Model (RoVRM) which improves human-preference alignment for LVLMs. RoVRM leverages auxiliary textual preference data through a three-phase progressive training and optimal transport-based preference data selection to effectively mitigate the scarcity of visual preference data. We experiment with RoVRM on the commonly used vision-language tasks based on the LLaVA-1.5-7B and -13B models. Experimental results demonstrate that RoVRM consistently outperforms traditional VRMs. Furthermore, our three-phase progressive training and preference data selection approaches can yield consistent performance gains over ranking-based alignment techniques, such as direct preference optimization.","url_abs":"https://arxiv.org/abs/2408.12109v2","url_pdf":"https://arxiv.org/pdf/2408.12109v2.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":"rovrm-a-robust-visual-reward-model-optimized","repo_url":"https://github.com/wangclnlp/vision-llm-alignment","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"hallucination","task_name":"Hallucination"}],"methods":[{"method_slug":"align","method_name":"ALIGN"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2408.12109","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2408.12109"}},"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/wangclnlp/vision-llm-alignment","reach":{"status":"ok"}}],"summary":{"ran":3,"ran_honours":1,"unverified":2},"by_repo_kind":{"official":{"samples":6,"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":6,"samples":[{"code_sha256_prefix":"ae99b6b77f74c9ae","entry":"get_eval_ds_config","repo":"wangclnlp/vision-llm-alignment","repo_kind":"official","path":"training/utils/ds_utils.py","file_url":"https://github.com/wangclnlp/vision-llm-alignment/blob/HEAD/training/utils/ds_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ae99b6b77f74c9ae"}},{"code_sha256_prefix":"9d10b3b5c4e7a742","entry":"get_train_ds_config","repo":"wangclnlp/vision-llm-alignment","repo_kind":"official","path":"training/utils/ds_utils.py","file_url":"https://github.com/wangclnlp/vision-llm-alignment/blob/HEAD/training/utils/ds_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9d10b3b5c4e7a742"}},{"code_sha256_prefix":"9b3c1cb391672ccb","entry":"load_image","repo":"wangclnlp/vision-llm-alignment","repo_kind":"official","path":"eval/predict_llava.py","file_url":"https://github.com/wangclnlp/vision-llm-alignment/blob/HEAD/eval/predict_llava.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"9b3c1cb391672ccb"}},{"code_sha256_prefix":"39d90ae4f80dca84","entry":"sampling_llama","repo":"wangclnlp/vision-llm-alignment","repo_kind":"official","path":"training/ppo_training/ppo_training_utils.py","file_url":"https://github.com/wangclnlp/vision-llm-alignment/blob/HEAD/training/ppo_training/ppo_training_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"39d90ae4f80dca84"}},{"code_sha256_prefix":"08e0a124270c05aa","entry":"sampling","repo":"wangclnlp/vision-llm-alignment","repo_kind":"official","path":"training/ppo_training/ppo_training_utils.py","file_url":"https://github.com/wangclnlp/vision-llm-alignment/blob/HEAD/training/ppo_training/ppo_training_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":"08e0a124270c05aa"}},{"code_sha256_prefix":"330f0a7bf45db596","entry":"sampling_llava","repo":"wangclnlp/vision-llm-alignment","repo_kind":"official","path":"training/ppo_training/ppo_training_utils.py","file_url":"https://github.com/wangclnlp/vision-llm-alignment/blob/HEAD/training/ppo_training/ppo_training_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":"330f0a7bf45db596"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}