{"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/vlfeedback-a-large-scale-ai-feedback-dataset","title":"VLFeedback: A Large-Scale AI Feedback Dataset for Large Vision-Language Models Alignment","arxiv_id":"2410.09421","date":"2024-10-12","proceeding":null,"authors":["Lei LI","Zhihui Xie","Mukai Li","Shunian Chen","Peiyi Wang","Liang Chen","Yazheng Yang","Benyou Wang","Lingpeng Kong","Qi Liu"],"abstract":"As large vision-language models (LVLMs) evolve rapidly, the demand for high-quality and diverse data to align these models becomes increasingly crucial. However, the creation of such data with human supervision proves costly and time-intensive. In this paper, we investigate the efficacy of AI feedback to scale supervision for aligning LVLMs. We introduce VLFeedback, the first large-scale vision-language feedback dataset, comprising over 82K multi-modal instructions and comprehensive rationales generated by off-the-shelf models without human annotations. To evaluate the effectiveness of AI feedback for vision-language alignment, we train Silkie, an LVLM fine-tuned via direct preference optimization on VLFeedback. Silkie showcases exceptional performance regarding helpfulness, visual faithfulness, and safety metrics. It outperforms its base model by 6.9\\% and 9.5\\% in perception and cognition tasks, reduces hallucination issues on MMHal-Bench, and exhibits enhanced resilience against red-teaming attacks. Furthermore, our analysis underscores the advantage of AI feedback, particularly in fostering preference diversity to deliver more comprehensive improvements. Our dataset, training code and models are available at https://vlf-silkie.github.io.","url_abs":"https://arxiv.org/abs/2410.09421v2","url_pdf":"https://arxiv.org/pdf/2410.09421v2.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":[],"tasks":[{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"hallucination","task_name":"Hallucination"},{"task_slug":"models-alignment","task_name":"Models Alignment"},{"task_slug":"red-teaming","task_name":"Red Teaming"},{"task_slug":"visual-question-answering-1","task_name":"Visual Question Answering"}],"methods":[{"method_slug":"align","method_name":"ALIGN"},{"method_slug":"base","method_name":"BASE"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/visual-question-answering-on-mm-vet","task":"Visual Question Answering","dataset":"MM-Vet","model":"Qwen-VL-Chat (+ SFT (GPT-4V in VLFeedback))","rank_in_archive_order":62,"of":231,"metrics":{"GPT-4 score":"50.7"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-mm-vet","task":"Visual Question Answering","dataset":"MM-Vet","model":"Silkie (Qwen-VL-Chat + DPO w/ VLFeedback)","rank_in_archive_order":67,"of":231,"metrics":{"GPT-4 score":"49.9"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-mm-vet","task":"Visual Question Answering","dataset":"MM-Vet","model":"LLaVA-Next-Mistral-7b (+ DPO w/ VLFeedback)","rank_in_archive_order":88,"of":231,"metrics":{"GPT-4 score":"44.2"},"uses_additional_data":false},{"leaderboard":"/sota/visual-question-answering-on-mm-vet","task":"Visual Question Answering","dataset":"MM-Vet","model":"LLaVA-Next-Vicuna-7b (+ DPO w/ VLFeedback)","rank_in_archive_order":90,"of":231,"metrics":{"GPT-4 score":"44.1"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2410.09421","atlas_url":"https://app.syntology.ai/?focus=2410.09421","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.09421"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/yuweihao/MM-Vet","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran_draft_wrong":5,"unverified":2},"by_repo_kind":{"found_in_text":{"samples":7,"ran":5,"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":"f41cb1a19b154297","entry":"encode_image","repo":"yuweihao/MM-Vet","repo_kind":"found_in_text","path":"inference/utils.py","file_url":"https://github.com/yuweihao/MM-Vet/blob/HEAD/inference/utils.py","link_basis":"harvester_set","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":"f41cb1a19b154297"}},{"code_sha256_prefix":"7d2672b31e9aa80e","entry":"get_file_names","repo":"yuweihao/MM-Vet","repo_kind":"found_in_text","path":"mm-vet_evaluator.py","file_url":"https://github.com/yuweihao/MM-Vet/blob/HEAD/mm-vet_evaluator.py","link_basis":"plan_row","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":"7d2672b31e9aa80e"}},{"code_sha256_prefix":"a907f50fb10a0162","entry":"get_file_names","repo":"yuweihao/MM-Vet","repo_kind":"found_in_text","path":"v2/mm-vet-v2_evaluator.py","file_url":"https://github.com/yuweihao/MM-Vet/blob/HEAD/v2/mm-vet-v2_evaluator.py","link_basis":"plan_row","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":"a907f50fb10a0162"}},{"code_sha256_prefix":"969f70cdb9e7b42d","entry":"load_metadata","repo":"yuweihao/MM-Vet","repo_kind":"found_in_text","path":"mm-vet_evaluator.py","file_url":"https://github.com/yuweihao/MM-Vet/blob/HEAD/mm-vet_evaluator.py","link_basis":"plan_row","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":"969f70cdb9e7b42d"}},{"code_sha256_prefix":"9385df01fd8cdfee","entry":"load_metadata","repo":"yuweihao/MM-Vet","repo_kind":"found_in_text","path":"v2/mm-vet-v2_evaluator.py","file_url":"https://github.com/yuweihao/MM-Vet/blob/HEAD/v2/mm-vet-v2_evaluator.py","link_basis":"plan_row","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":"9385df01fd8cdfee"}},{"code_sha256_prefix":"4edc6079065ecd0f","entry":"arg_parser","repo":"yuweihao/MM-Vet","repo_kind":"found_in_text","path":"mm-vet_evaluator.py","file_url":"https://github.com/yuweihao/MM-Vet/blob/HEAD/mm-vet_evaluator.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4edc6079065ecd0f"}},{"code_sha256_prefix":"436825beacc9df7d","entry":"arg_parser","repo":"yuweihao/MM-Vet","repo_kind":"found_in_text","path":"v2/mm-vet-v2_evaluator.py","file_url":"https://github.com/yuweihao/MM-Vet/blob/HEAD/v2/mm-vet-v2_evaluator.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"436825beacc9df7d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}