{"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/kebench-a-benchmark-on-knowledge-editing-for","title":"VLKEB: A Large Vision-Language Model Knowledge Editing Benchmark","arxiv_id":"2403.07350","date":"2024-03-12","proceeding":null,"authors":["Han Huang","Haitian Zhong","Tao Yu","Qiang Liu","Shu Wu","Liang Wang","Tieniu Tan"],"abstract":"Recently, knowledge editing on large language models (LLMs) has received considerable attention. Compared to this, editing Large Vision-Language Models (LVLMs) faces extra challenges from diverse data modalities and complicated model components, and data for LVLMs editing are limited. The existing LVLM editing benchmark, which comprises three metrics (Reliability, Locality, and Generality), falls short in the quality of synthesized evaluation images and cannot assess whether models apply edited knowledge in relevant content. Therefore, we employ more reliable data collection methods to construct a new Large $\\textbf{V}$ision-$\\textbf{L}$anguage Model $\\textbf{K}$nowledge $\\textbf{E}$diting $\\textbf{B}$enchmark, $\\textbf{VLKEB}$, and extend the Portability metric for more comprehensive evaluation. Leveraging a multi-modal knowledge graph, our image data are bound with knowledge entities. This can be further used to extract entity-related knowledge, which constitutes the base of editing data. We conduct experiments of different editing methods on five LVLMs, and thoroughly analyze how do they impact the models. The results reveal strengths and deficiencies of these methods and hopefully provide insights for future research. The codes and dataset are available at: https://github.com/VLKEB/VLKEB.","url_abs":"https://arxiv.org/abs/2403.07350v3","url_pdf":"https://arxiv.org/pdf/2403.07350v3.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":"kebench-a-benchmark-on-knowledge-editing-for","repo_url":"https://github.com/vlkeb/vlkeb","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"multi-modal-knowledge-graph","task_name":"Multi-modal Knowledge Graph"},{"task_slug":"knowledge-editing","task_name":"knowledge editing"}],"methods":[{"method_slug":"mend","method_name":"MEND"},{"method_slug":"serac","method_name":"SERAC"}],"datasets_introduced":[{"slug":"vlkeb","name":"VLKEB","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2403.07350","atlas_url":"https://app.syntology.ai/?focus=2403.07350","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.07350"}},"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/VLKEB/VLKEB","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/X-PLUG/mPLUG-Owl","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/vlkeb/vlkeb","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":14,"ran_draft_wrong":4,"ran_fixture":2,"unverified":5},"by_repo_kind":{"official":{"samples":12,"ran":8,"repositories":1},"found_in_text":{"samples":13,"ran":12,"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":"a3aac05439340a66","entry":"batchify","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/pipeline/utils.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/pipeline/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a3aac05439340a66"}},{"code_sha256_prefix":"27659a9c234ffb11","entry":"binary_log_probs","repo":"VLKEB/VLKEB","repo_kind":"official","path":"easyeditor/trainer/losses.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/easyeditor/trainer/losses.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":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"27659a9c234ffb11"}},{"code_sha256_prefix":"1a531e3c0d1094fb","entry":"bloom_forward","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/mplug_owl/modeling_mplug_owl.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/mplug_owl/modeling_mplug_owl.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1a531e3c0d1094fb"}},{"code_sha256_prefix":"3f4963e78e50be03","entry":"chunk_it","repo":"VLKEB/VLKEB","repo_kind":"official","path":"KE/src/utils.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/KE/src/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3f4963e78e50be03"}},{"code_sha256_prefix":"45711ccd8932d38a","entry":"detokenize_generations","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/mplug_owl/processing_mplug_owl.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/mplug_owl/processing_mplug_owl.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"45711ccd8932d38a"}},{"code_sha256_prefix":"7a905b93523a9f85","entry":"do_generate","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/pipeline/interface.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/pipeline/interface.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7a905b93523a9f85"}},{"code_sha256_prefix":"97b29adf427aa64e","entry":"get_cosine_schedule_with_warmup","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/pipeline/utils.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/pipeline/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"97b29adf427aa64e"}},{"code_sha256_prefix":"45567e61fe09093a","entry":"get_ltor_masks_and_position_ids_from_embeddings","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/mplug_owl/modeling_mplug_owl.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/mplug_owl/modeling_mplug_owl.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"45567e61fe09093a"}},{"code_sha256_prefix":"265f4d55ccd60d18","entry":"get_media_indices","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/mplug_owl/modeling_mplug_owl.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/mplug_owl/modeling_mplug_owl.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"265f4d55ccd60d18"}},{"code_sha256_prefix":"752b84a6f01c43d8","entry":"get_media_types","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/mplug_owl_video/modeling_mplug_owl.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/mplug_owl_video/modeling_mplug_owl.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"752b84a6f01c43d8"}},{"code_sha256_prefix":"53b3ed9d6fed2b5f","entry":"get_param_groups","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/pipeline/utils.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/pipeline/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"53b3ed9d6fed2b5f"}},{"code_sha256_prefix":"920b394e7ad80c53","entry":"hierarchical_subsequence","repo":"VLKEB/VLKEB","repo_kind":"official","path":"easyeditor/util/nethook.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/easyeditor/util/nethook.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"920b394e7ad80c53"}},{"code_sha256_prefix":"96b6713ff1c322ca","entry":"label_smoothed_nll_loss","repo":"VLKEB/VLKEB","repo_kind":"official","path":"KE/src/utils.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/KE/src/utils.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"96b6713ff1c322ca"}},{"code_sha256_prefix":"4609b1b3a11479ce","entry":"multiclass_log_probs","repo":"VLKEB/VLKEB","repo_kind":"official","path":"easyeditor/trainer/losses.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/easyeditor/trainer/losses.py","link_basis":"plan_row","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4609b1b3a11479ce"}},{"code_sha256_prefix":"23d3acf0aced66d5","entry":"normalize","repo":"VLKEB/VLKEB","repo_kind":"official","path":"KE/src/utils.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/KE/src/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"23d3acf0aced66d5"}},{"code_sha256_prefix":"15e54d5d53ab90ac","entry":"post_process_code","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/serve/model_utils.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/serve/model_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"15e54d5d53ab90ac"}},{"code_sha256_prefix":"607a861da4671ce5","entry":"post_process_output","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/serve/model_utils.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/serve/model_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"607a861da4671ce5"}},{"code_sha256_prefix":"70f6ab8bde55420e","entry":"recursive_copy","repo":"VLKEB/VLKEB","repo_kind":"official","path":"easyeditor/util/nethook.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/easyeditor/util/nethook.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"70f6ab8bde55420e"}},{"code_sha256_prefix":"440ff98c2b1ae1aa","entry":"subsequence","repo":"VLKEB/VLKEB","repo_kind":"official","path":"easyeditor/util/nethook.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/easyeditor/util/nethook.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"440ff98c2b1ae1aa"}},{"code_sha256_prefix":"a314d930b7b0cfbb","entry":"tokenize_prompts","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/mplug_owl/processing_mplug_owl.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/mplug_owl/processing_mplug_owl.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a314d930b7b0cfbb"}},{"code_sha256_prefix":"65d587525000912a","entry":"get_index","repo":"X-PLUG/mPLUG-Owl","repo_kind":"found_in_text","path":"mPLUG-Owl/mplug_owl_video/processing_mplug_owl.py","file_url":"https://github.com/X-PLUG/mPLUG-Owl/blob/HEAD/mPLUG-Owl/mplug_owl_video/processing_mplug_owl.py","link_basis":"plan_row","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"65d587525000912a"}},{"code_sha256_prefix":"a0e211c2cfa8303a","entry":"get_model","repo":"VLKEB/VLKEB","repo_kind":"official","path":"easyeditor/trainer/models.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/easyeditor/trainer/models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a0e211c2cfa8303a"}},{"code_sha256_prefix":"560e686f4709705f","entry":"get_model","repo":"VLKEB/VLKEB","repo_kind":"official","path":"KE/src/models/get_models.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/KE/src/models/get_models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"560e686f4709705f"}},{"code_sha256_prefix":"77ff1d0bef624163","entry":"get_tokenizer","repo":"VLKEB/VLKEB","repo_kind":"official","path":"easyeditor/trainer/models.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/easyeditor/trainer/models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"77ff1d0bef624163"}},{"code_sha256_prefix":"bcc592aa5c1e2bb3","entry":"kl_loc_loss","repo":"VLKEB/VLKEB","repo_kind":"official","path":"easyeditor/trainer/losses.py","file_url":"https://github.com/VLKEB/VLKEB/blob/HEAD/easyeditor/trainer/losses.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"bcc592aa5c1e2bb3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}