{"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/surf-teaching-large-vision-language-models-to","title":"SURf: Teaching Large Vision-Language Models to Selectively Utilize Retrieved Information","arxiv_id":"2409.14083","date":"2024-09-21","proceeding":null,"authors":["Jiashuo Sun","Jihai Zhang","Yucheng Zhou","Zhaochen Su","Xiaoye Qu","Yu Cheng"],"abstract":"Large Vision-Language Models (LVLMs) have become pivotal at the intersection of computer vision and natural language processing. However, the full potential of LVLMs Retrieval-Augmented Generation (RAG) capabilities remains underutilized. Existing works either focus solely on the text modality or are limited to specific tasks. Moreover, most LVLMs struggle to selectively utilize retrieved information and are sensitive to irrelevant or misleading references. To address these challenges, we propose a self-refinement framework designed to teach LVLMs to Selectively Utilize Retrieved Information (SURf). Specifically, when given questions that are incorrectly answered by the LVLM backbone, we obtain references that help correct the answers (positive references) and those that do not (negative references). We then fine-tune the LVLM backbone using a combination of these positive and negative references. Our experiments across three tasks and seven datasets demonstrate that our framework significantly enhances LVLMs ability to effectively utilize retrieved multimodal references and improves their robustness against irrelevant or misleading information. The source code is available at https://github.com/GasolSun36/SURf.","url_abs":"https://arxiv.org/abs/2409.14083v1","url_pdf":"https://arxiv.org/pdf/2409.14083v1.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":"surf-teaching-large-vision-language-models-to","repo_url":"https://github.com/gasolsun36/surf","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"rag","task_name":"RAG"},{"task_slug":"retrieval-augmented-generation","task_name":"Retrieval-augmented Generation"}],"methods":[{"method_slug":"focus","method_name":"Focus"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2409.14083","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2409.14083"}},"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/gasolsun36/surf","reach":{"status":"ok"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/GasolSun36/SURf","reach":{"status":"ok"}}],"summary":{"ran":15,"ran_draft_wrong":2,"ran_fixture":1},"by_repo_kind":{"official":{"samples":18,"ran":18,"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":18,"samples":[{"code_sha256_prefix":"c8c51dd88ad58cb1","entry":"build_prompt","repo":"GasolSun36/SURf","repo_kind":"official","path":"eval/pope.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/eval/pope.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c8c51dd88ad58cb1"}},{"code_sha256_prefix":"48f2a9620f2ac197","entry":"build_prompt","repo":"GasolSun36/SURf","repo_kind":"official","path":"initial/data_filter.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/initial/data_filter.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"48f2a9620f2ac197"}},{"code_sha256_prefix":"31fd3eb13a63be04","entry":"calculate_metrics","repo":"GasolSun36/SURf","repo_kind":"official","path":"eval/eval_pope.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/eval/eval_pope.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":"31fd3eb13a63be04"}},{"code_sha256_prefix":"f63fb039437be784","entry":"compute_statistics","repo":"GasolSun36/SURf","repo_kind":"official","path":"eval/eval_pope.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/eval/eval_pope.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":"f63fb039437be784"}},{"code_sha256_prefix":"4267a0b07402b7e7","entry":"create_faiss_index","repo":"GasolSun36/SURf","repo_kind":"official","path":"database/create_index.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/database/create_index.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":"4267a0b07402b7e7"}},{"code_sha256_prefix":"7e03b180fa317c9a","entry":"divide_to_patches","repo":"GasolSun36/SURf","repo_kind":"official","path":"llava/mm_utils.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/llava/mm_utils.py","link_basis":"harvester_set","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":"7e03b180fa317c9a"}},{"code_sha256_prefix":"442f044af42b5b3f","entry":"exact_match","repo":"GasolSun36/SURf","repo_kind":"official","path":"initial/tool_evaluate.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/initial/tool_evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"442f044af42b5b3f"}},{"code_sha256_prefix":"94855420b055f5cb","entry":"generate_clip_vectors","repo":"GasolSun36/SURf","repo_kind":"official","path":"database/create_index.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/database/create_index.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":"94855420b055f5cb"}},{"code_sha256_prefix":"42a46570620cd9fa","entry":"get_chunk","repo":"GasolSun36/SURf","repo_kind":"official","path":"initial/generate_initial_data.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/initial/generate_initial_data.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"well_formed","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"42a46570620cd9fa"}},{"code_sha256_prefix":"06d40f4711e62357","entry":"get_chunk","repo":"GasolSun36/SURf","repo_kind":"official","path":"eval/pope.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/eval/pope.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"06d40f4711e62357"}},{"code_sha256_prefix":"8a203abf91abfcad","entry":"get_chunk","repo":"gasolsun36/surf","repo_kind":"official","path":"initial/tool_evaluate.py","file_url":"https://github.com/gasolsun36/surf/blob/HEAD/initial/tool_evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8a203abf91abfcad"}},{"code_sha256_prefix":"c440a6c78fc9fc04","entry":"load_faiss_index","repo":"GasolSun36/SURf","repo_kind":"official","path":"eval/pope.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/eval/pope.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":"c440a6c78fc9fc04"}},{"code_sha256_prefix":"1bdb9a6725f61c1d","entry":"load_files","repo":"GasolSun36/SURf","repo_kind":"official","path":"eval/eval_pope.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/eval/eval_pope.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":"1bdb9a6725f61c1d"}},{"code_sha256_prefix":"468eedeba67f1b00","entry":"resize_and_pad_image","repo":"GasolSun36/SURf","repo_kind":"official","path":"llava/mm_utils.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/llava/mm_utils.py","link_basis":"harvester_set","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":"468eedeba67f1b00"}},{"code_sha256_prefix":"e6cc6a9ac7adfd83","entry":"retrieve_base_filenames","repo":"GasolSun36/SURf","repo_kind":"official","path":"initial/data_filter.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/initial/data_filter.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":"e6cc6a9ac7adfd83"}},{"code_sha256_prefix":"3999ff487573f32c","entry":"select_best_resolution","repo":"GasolSun36/SURf","repo_kind":"official","path":"llava/mm_utils.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/llava/mm_utils.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3999ff487573f32c"}},{"code_sha256_prefix":"076c252c52cbb161","entry":"split_list","repo":"GasolSun36/SURf","repo_kind":"official","path":"initial/generate_initial_data.py","file_url":"https://github.com/GasolSun36/SURf/blob/HEAD/initial/generate_initial_data.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"076c252c52cbb161"}},{"code_sha256_prefix":"abe4f5d245fa207a","entry":"split_list","repo":"gasolsun36/surf","repo_kind":"official","path":"initial/tool_evaluate.py","file_url":"https://github.com/gasolsun36/surf/blob/HEAD/initial/tool_evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"abe4f5d245fa207a"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}