{"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/woodpecker-hallucination-correction-for","title":"Woodpecker: Hallucination Correction for Multimodal Large Language Models","arxiv_id":"2310.16045","date":"2023-10-24","proceeding":null,"authors":["Shukang Yin","Chaoyou Fu","Sirui Zhao","Tong Xu","Hao Wang","Dianbo Sui","Yunhang Shen","Ke Li","Xing Sun","Enhong Chen"],"abstract":"Hallucination is a big shadow hanging over the rapidly evolving Multimodal Large Language Models (MLLMs), referring to the phenomenon that the generated text is inconsistent with the image content. In order to mitigate hallucinations, existing studies mainly resort to an instruction-tuning manner that requires retraining the models with specific data. In this paper, we pave a different way, introducing a training-free method named Woodpecker. Like a woodpecker heals trees, it picks out and corrects hallucinations from the generated text. Concretely, Woodpecker consists of five stages: key concept extraction, question formulation, visual knowledge validation, visual claim generation, and hallucination correction. Implemented in a post-remedy manner, Woodpecker can easily serve different MLLMs, while being interpretable by accessing intermediate outputs of the five stages. We evaluate Woodpecker both quantitatively and qualitatively and show the huge potential of this new paradigm. On the POPE benchmark, our method obtains a 30.66%/24.33% improvement in accuracy over the baseline MiniGPT-4/mPLUG-Owl. The source code is released at https://github.com/BradyFU/Woodpecker.","url_abs":"https://arxiv.org/abs/2310.16045v2","url_pdf":"https://arxiv.org/pdf/2310.16045v2.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":"woodpecker-hallucination-correction-for","repo_url":"https://github.com/bradyfu/woodpecker","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"hallucination","task_name":"Hallucination"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2310.16045","atlas_url":"https://app.syntology.ai/?focus=2310.16045","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.16045"}},"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":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/bradyfu/woodpecker","reach":{"status":"ok"}}],"summary":{"ran":7,"unverified":6},"by_repo_kind":{"official":{"samples":13,"ran":7,"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":13,"samples":[{"code_sha256_prefix":"667ecbfb72739567","entry":"compute_iou","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/utils.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/utils.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":"667ecbfb72739567"}},{"code_sha256_prefix":"ded3d03af7363e20","entry":"extract_boxes","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/utils.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/utils.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":"ded3d03af7363e20"}},{"code_sha256_prefix":"b5121592afbce3f8","entry":"find_matching_boxes","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/utils.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/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":"b5121592afbce3f8"}},{"code_sha256_prefix":"d1ffbb6af3208715","entry":"format_inputs","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/claim_generator.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/claim_generator.py","link_basis":"harvester_set","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":"d1ffbb6af3208715"}},{"code_sha256_prefix":"1e1856185896a734","entry":"get_claim","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/claim_generator.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/claim_generator.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":"1e1856185896a734"}},{"code_sha256_prefix":"71e7ce8bf939f5d4","entry":"get_res","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/questioner.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/questioner.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":"71e7ce8bf939f5d4"}},{"code_sha256_prefix":"8a2ce3b99884243a","entry":"remove_duplicates","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/questioner.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/questioner.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":"8a2ce3b99884243a"}},{"code_sha256_prefix":"9528b19eb2cc3b97","entry":"find_most_similar_strings","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/detector.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/detector.py","link_basis":"harvester_set","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":"9528b19eb2cc3b97"}},{"code_sha256_prefix":"9698cd221b31fdd5","entry":"get_all_answers","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/answerer.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/answerer.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":"9698cd221b31fdd5"}},{"code_sha256_prefix":"20117a653c8f40ac","entry":"get_answer","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/answerer.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/answerer.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":"20117a653c8f40ac"}},{"code_sha256_prefix":"8122965ad6131653","entry":"get_output","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/preprocessor.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/preprocessor.py","link_basis":"harvester_set","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":"8122965ad6131653"}},{"code_sha256_prefix":"3cc2c9efff7e5609","entry":"get_output","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/refiner.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/refiner.py","link_basis":"harvester_set","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":"3cc2c9efff7e5609"}},{"code_sha256_prefix":"8dbb0079cf7049ec","entry":"get_res","repo":"bradyfu/woodpecker","repo_kind":"official","path":"models/entity_extractor.py","file_url":"https://github.com/bradyfu/woodpecker/blob/HEAD/models/entity_extractor.py","link_basis":"harvester_set","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":"8dbb0079cf7049ec"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}