{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:2MD3PLBMT6QKU3XL2GAJZJJ6B7","short_pith_number":"pith:2MD3PLBM","schema_version":"1.0","canonical_sha256":"d307b7ac2c9fa0aa6eebd1809ca53e0fc7d15b6e9b301c0acabd365aab3b8676","source":{"kind":"arxiv","id":"2307.16125","version":2},"attestation_state":"computed","paper":{"title":"SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension","license":"http://creativecommons.org/licenses/by/4.0/","headline":"SEED-Bench supplies 19K human-verified multiple-choice questions to measure multimodal LLMs on image and video comprehension across 12 dimensions.","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Bohao Li, Guangzhi Wang, Rui Wang, Ying Shan, Yixiao Ge, Yuying Ge","submitted_at":"2023-07-30T04:25:16Z","abstract_excerpt":"Based on powerful Large Language Models (LLMs), recent generative Multimodal Large Language Models (MLLMs) have gained prominence as a pivotal research area, exhibiting remarkable capability for both comprehension and generation. In this work, we address the evaluation of generative comprehension in MLLMs as a preliminary step towards a comprehensive assessment of generative models, by introducing a benchmark named SEED-Bench. SEED-Bench consists of 19K multiple choice questions with accurate human annotations (x 6 larger than existing benchmarks), which spans 12 evaluation dimensions includin"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":true,"formal_links_present":true},"canonical_record":{"source":{"id":"2307.16125","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-30T04:25:16Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"3cfa260a255e77b76cec5139aa572843eb9b952de93949abd296a45d044f1902","abstract_canon_sha256":"8b5fdc45f6d246ff05679d76140b94cc8ef1bd3b82a6eaf00c67fead7297a2a5"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:36:55.687271Z","signature_b64":"uoAcFJCiwiu9RDlRt6NhTysVs49Xko05wetSnKt/qZVR09miGC6JEKeBiJxgyL4CYD2g6YN4aT2/XyU+c9eoAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d307b7ac2c9fa0aa6eebd1809ca53e0fc7d15b6e9b301c0acabd365aab3b8676","last_reissued_at":"2026-07-05T06:36:55.686764Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:36:55.686764Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension","license":"http://creativecommons.org/licenses/by/4.0/","headline":"SEED-Bench supplies 19K human-verified multiple-choice questions to measure multimodal LLMs on image and video comprehension across 12 dimensions.","cross_cats":["cs.CV"],"primary_cat":"cs.CL","authors_text":"Bohao Li, Guangzhi Wang, Rui Wang, Ying Shan, Yixiao Ge, Yuying Ge","submitted_at":"2023-07-30T04:25:16Z","abstract_excerpt":"Based on powerful Large Language Models (LLMs), recent generative Multimodal Large Language Models (MLLMs) have gained prominence as a pivotal research area, exhibiting remarkable capability for both comprehension and generation. In this work, we address the evaluation of generative comprehension in MLLMs as a preliminary step towards a comprehensive assessment of generative models, by introducing a benchmark named SEED-Bench. SEED-Bench consists of 19K multiple choice questions with accurate human annotations (x 6 larger than existing benchmarks), which spans 12 evaluation dimensions includin"},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"SEED-Bench consists of 19K multiple choice questions with accurate human annotations (x 6 larger than existing benchmarks), which spans 12 evaluation dimensions including the comprehension of both the image and video modality.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the pipeline of automatic question generation plus manual verification produces questions that genuinely measure generative comprehension rather than artifacts of the generation process itself.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"SEED-Bench is a new benchmark of 19K multiple-choice questions for evaluating generative comprehension in multimodal LLMs across 12 image and video dimensions.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"SEED-Bench supplies 19K human-verified multiple-choice questions to measure multimodal LLMs on image and video comprehension across 12 dimensions.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"446d3742bfce8b2df877ed41886b90a9833d90d6939bcd7a74319525b99f20e2"},"source":{"id":"2307.16125","kind":"arxiv","version":2},"verdict":{"id":"dede37a3-26a2-4e9a-870a-4cc4cef5795b","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-12T16:55:12.675278Z","strongest_claim":"SEED-Bench consists of 19K multiple choice questions with accurate human annotations (x 6 larger than existing benchmarks), which spans 12 evaluation dimensions including the comprehension of both the image and video modality.","one_line_summary":"SEED-Bench is a new benchmark of 19K multiple-choice questions for evaluating generative comprehension in multimodal LLMs across 12 image and video dimensions.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the pipeline of automatic question generation plus manual verification produces questions that genuinely measure generative comprehension rather than artifacts of the generation process itself.","pith_extraction_headline":"SEED-Bench supplies 19K human-verified multiple-choice questions to measure multimodal LLMs on image and video comprehension across 12 dimensions."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2307.16125/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":42,"sample":[{"doi":"","year":2022,"title":"Scaling Instruction-Finetuned Language Models","work_id":"8405abb1-7558-4fdf-af24-f4c52fa77a06","ref_index":1,"cited_arxiv_id":"2210.11416","is_internal_anchor":true},{"doi":"","year":2023,"title":"Gpt-4 technical report","work_id":"388f534c-855a-4366-b933-f07bf3e2db5f","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2022,"title":"Introducing chatgpt","work_id":"1b840f0b-fb6b-43d7-bc36-05b0a98a171e","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"FastChat. Vicuna. https://github.com/lm-sys/FastChat, 2023","work_id":"ee2e3957-facb-454c-b74c-07fdc354f73e","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2023,"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","ref_index":5,"cited_arxiv_id":"2302.13971","is_internal_anchor":true}],"resolved_work":42,"snapshot_sha256":"b967801c3b760b0eff858d80d9e18d6823f2cfd7743a7f5011bd81be2bf3f7e3","internal_anchors":16},"formal_canon":{"evidence_count":2,"snapshot_sha256":"488a2e33743f22e32483cda8e09b1b26e774339717002e8ebfc44bb5f09d6748"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2307.16125","created_at":"2026-07-05T06:36:55.686828+00:00"},{"alias_kind":"arxiv_version","alias_value":"2307.16125v2","created_at":"2026-07-05T06:36:55.686828+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.16125","created_at":"2026-07-05T06:36:55.686828+00:00"},{"alias_kind":"pith_short_12","alias_value":"2MD3PLBMT6QK","created_at":"2026-07-05T06:36:55.686828+00:00"},{"alias_kind":"pith_short_16","alias_value":"2MD3PLBMT6QKU3XL","created_at":"2026-07-05T06:36:55.686828+00:00"},{"alias_kind":"pith_short_8","alias_value":"2MD3PLBM","created_at":"2026-07-05T06:36:55.686828+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":117,"internal_anchor_count":117,"sample":[{"citing_arxiv_id":"2606.26029","citing_title":"TriViewBench: Controlled Complexity Scaling for Multi-View Structural Reasoning in MLLMs","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2606.25034","citing_title":"Yuvion VL: A Multimodal Foundation Model for Adversarial Content and AI Safety","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2606.23885","citing_title":"Mind the Heads: Topological Representation Alignment for Multimodal LLMs","ref_index":16,"is_internal_anchor":true},{"citing_arxiv_id":"2606.21734","citing_title":"HPP: Hierarchical Programmatic Probing for Long Video Understanding by Decoupling Perception and Reasoning","ref_index":170,"is_internal_anchor":true},{"citing_arxiv_id":"2606.19706","citing_title":"NEST: Narrative Event Structures in Time for Long Video Understanding","ref_index":290,"is_internal_anchor":true},{"citing_arxiv_id":"2606.20077","citing_title":"The Hidden Evolution of Disguised Visual Context inside the VLM","ref_index":33,"is_internal_anchor":true},{"citing_arxiv_id":"2606.28864","citing_title":"On Test-Time Scaling for Vision-Language Models","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2606.17389","citing_title":"Visuals Lie, Consistency Speaks: Disentangling Spatial Attention from Reliability in Vision-Language Models","ref_index":4,"is_internal_anchor":true},{"citing_arxiv_id":"2606.17030","citing_title":"Qwen-RobotWorld Technical Report: Unifying Embodied World Modeling through Language-Conditioned Video Generation","ref_index":72,"is_internal_anchor":true},{"citing_arxiv_id":"2606.13289","citing_title":"HYDRA-X: Native Unified Multimodal Models with Holistic Visual Tokenizers","ref_index":211,"is_internal_anchor":true},{"citing_arxiv_id":"2606.12744","citing_title":"GRIP: Feedback-Guided Prompt Retrieval for Large Multimodal Models","ref_index":7,"is_internal_anchor":true},{"citing_arxiv_id":"2606.12412","citing_title":"Reroute, Don't Remove: Recoverable Visual Token Routing for Vision-Language Models","ref_index":34,"is_internal_anchor":true},{"citing_arxiv_id":"2606.11074","citing_title":"Modeling Complex Behaviors: Multi-Personality Composition and Dynamic Switching in Vision-Language Models","ref_index":78,"is_internal_anchor":true},{"citing_arxiv_id":"2606.11096","citing_title":"IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder","ref_index":24,"is_internal_anchor":true},{"citing_arxiv_id":"2606.09393","citing_title":"CapRL++: Unified Reinforcement Learning with Verifiable Rewards for Dense Image and Video Captioning","ref_index":36,"is_internal_anchor":true},{"citing_arxiv_id":"2606.28551","citing_title":"DataComp-VLM: Improved Open Datasets for Vision-Language Models","ref_index":148,"is_internal_anchor":true},{"citing_arxiv_id":"2606.07861","citing_title":"The Last Visible Pixel: Probing Fine-Scale Perception in Vision-Language Models","ref_index":27,"is_internal_anchor":true},{"citing_arxiv_id":"2606.06853","citing_title":"MotionEnhancer: Leveraging Video Diffusion for Motion-Enhanced Vision-Language Models","ref_index":12,"is_internal_anchor":true},{"citing_arxiv_id":"2606.06890","citing_title":"Diagnosing Visual Ignorance in Vision-Language Models","ref_index":25,"is_internal_anchor":true},{"citing_arxiv_id":"2606.05232","citing_title":"Differentiable Efficient Operator Search","ref_index":8,"is_internal_anchor":true},{"citing_arxiv_id":"2606.07653","citing_title":"A Dataset for Dynamic Human Preferences for Vision Language Models","ref_index":18,"is_internal_anchor":true},{"citing_arxiv_id":"2606.07643","citing_title":"AVI-Bench: Toward Human-like Audio-Visual Intelligence of Omni-MLLMs","ref_index":65,"is_internal_anchor":true},{"citing_arxiv_id":"2606.07639","citing_title":"MOSS-Video-Preview: Toward Real-Time Video Understanding via Cross-Attention","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2606.01756","citing_title":"EvoCut: Multi-Layer Evolution-Aware Visual Token Compression for Efficient Large Vision-Language Models","ref_index":21,"is_internal_anchor":true},{"citing_arxiv_id":"2606.00390","citing_title":"Zamba2-VL Technical Report","ref_index":110,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":2,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2MD3PLBMT6QKU3XL2GAJZJJ6B7","json":"https://pith.science/pith/2MD3PLBMT6QKU3XL2GAJZJJ6B7.json","graph_json":"https://pith.science/api/pith-number/2MD3PLBMT6QKU3XL2GAJZJJ6B7/graph.json","events_json":"https://pith.science/api/pith-number/2MD3PLBMT6QKU3XL2GAJZJJ6B7/events.json","paper":"https://pith.science/paper/2MD3PLBM"},"agent_actions":{"view_html":"https://pith.science/pith/2MD3PLBMT6QKU3XL2GAJZJJ6B7","download_json":"https://pith.science/pith/2MD3PLBMT6QKU3XL2GAJZJJ6B7.json","view_paper":"https://pith.science/paper/2MD3PLBM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2307.16125&json=true","fetch_graph":"https://pith.science/api/pith-number/2MD3PLBMT6QKU3XL2GAJZJJ6B7/graph.json","fetch_events":"https://pith.science/api/pith-number/2MD3PLBMT6QKU3XL2GAJZJJ6B7/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2MD3PLBMT6QKU3XL2GAJZJJ6B7/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2MD3PLBMT6QKU3XL2GAJZJJ6B7/action/storage_attestation","attest_author":"https://pith.science/pith/2MD3PLBMT6QKU3XL2GAJZJJ6B7/action/author_attestation","sign_citation":"https://pith.science/pith/2MD3PLBMT6QKU3XL2GAJZJJ6B7/action/citation_signature","submit_replication":"https://pith.science/pith/2MD3PLBMT6QKU3XL2GAJZJJ6B7/action/replication_record"}},"created_at":"2026-07-05T06:36:55.686828+00:00","updated_at":"2026-07-05T06:36:55.686828+00:00"}