{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:2MD3PLBMT6QKU3XL2GAJZJJ6B7","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"8b5fdc45f6d246ff05679d76140b94cc8ef1bd3b82a6eaf00c67fead7297a2a5","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-30T04:25:16Z","title_canon_sha256":"3cfa260a255e77b76cec5139aa572843eb9b952de93949abd296a45d044f1902"},"schema_version":"1.0","source":{"id":"2307.16125","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2307.16125","created_at":"2026-07-05T06:36:55Z"},{"alias_kind":"arxiv_version","alias_value":"2307.16125v2","created_at":"2026-07-05T06:36:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2307.16125","created_at":"2026-07-05T06:36:55Z"},{"alias_kind":"pith_short_12","alias_value":"2MD3PLBMT6QK","created_at":"2026-07-05T06:36:55Z"},{"alias_kind":"pith_short_16","alias_value":"2MD3PLBMT6QKU3XL","created_at":"2026-07-05T06:36:55Z"},{"alias_kind":"pith_short_8","alias_value":"2MD3PLBM","created_at":"2026-07-05T06:36:55Z"}],"graph_snapshots":[{"event_id":"sha256:fc00873123b823ad76d09765c7108480619db9e3825469f3816817307e0f6eb7","target":"graph","created_at":"2026-07-05T06:36:55Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","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."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"SEED-Bench supplies 19K human-verified multiple-choice questions to measure multimodal LLMs on image and video comprehension across 12 dimensions."}],"snapshot_sha256":"446d3742bfce8b2df877ed41886b90a9833d90d6939bcd7a74319525b99f20e2"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"488a2e33743f22e32483cda8e09b1b26e774339717002e8ebfc44bb5f09d6748"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2307.16125/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Bohao Li, Guangzhi Wang, Rui Wang, Ying Shan, Yixiao Ge, Yuying Ge","cross_cats":["cs.CV"],"headline":"SEED-Bench supplies 19K human-verified multiple-choice questions to measure multimodal LLMs on image and video comprehension across 12 dimensions.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-30T04:25:16Z","title":"SEED-Bench: Benchmarking Multimodal LLMs with Generative Comprehension"},"references":{"count":42,"internal_anchors":16,"resolved_work":42,"sample":[{"cited_arxiv_id":"2210.11416","doi":"","is_internal_anchor":true,"ref_index":1,"title":"Scaling Instruction-Finetuned Language Models","work_id":"8405abb1-7558-4fdf-af24-f4c52fa77a06","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":2,"title":"Gpt-4 technical report","work_id":"388f534c-855a-4366-b933-f07bf3e2db5f","year":2023},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":3,"title":"Introducing chatgpt","work_id":"1b840f0b-fb6b-43d7-bc36-05b0a98a171e","year":2022},{"cited_arxiv_id":"","doi":"","is_internal_anchor":false,"ref_index":4,"title":"FastChat. Vicuna. https://github.com/lm-sys/FastChat, 2023","work_id":"ee2e3957-facb-454c-b74c-07fdc354f73e","year":2023},{"cited_arxiv_id":"2302.13971","doi":"","is_internal_anchor":true,"ref_index":5,"title":"LLaMA: Open and Efficient Foundation Language Models","work_id":"c018fc23-6f3f-4035-9d02-28a2173b2b9d","year":2023}],"snapshot_sha256":"b967801c3b760b0eff858d80d9e18d6823f2cfd7743a7f5011bd81be2bf3f7e3"},"source":{"id":"2307.16125","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-12T16:55:12.675278Z","id":"dede37a3-26a2-4e9a-870a-4cc4cef5795b","model_set":{"reader":"grok-4.3"},"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","pith_extraction_headline":"SEED-Bench supplies 19K human-verified multiple-choice questions to measure multimodal LLMs on image and video comprehension across 12 dimensions.","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.","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."}},"verdict_id":"dede37a3-26a2-4e9a-870a-4cc4cef5795b"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8afc525728495fa344a5d305da599ba5a7dbc93c24e3e1e593c0ff52c8e59ea0","target":"record","created_at":"2026-07-05T06:36:55Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"8b5fdc45f6d246ff05679d76140b94cc8ef1bd3b82a6eaf00c67fead7297a2a5","cross_cats_sorted":["cs.CV"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-07-30T04:25:16Z","title_canon_sha256":"3cfa260a255e77b76cec5139aa572843eb9b952de93949abd296a45d044f1902"},"schema_version":"1.0","source":{"id":"2307.16125","kind":"arxiv","version":2}},"canonical_sha256":"d307b7ac2c9fa0aa6eebd1809ca53e0fc7d15b6e9b301c0acabd365aab3b8676","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d307b7ac2c9fa0aa6eebd1809ca53e0fc7d15b6e9b301c0acabd365aab3b8676","first_computed_at":"2026-07-05T06:36:55.686764Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:36:55.686764Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uoAcFJCiwiu9RDlRt6NhTysVs49Xko05wetSnKt/qZVR09miGC6JEKeBiJxgyL4CYD2g6YN4aT2/XyU+c9eoAQ==","signature_status":"signed_v1","signed_at":"2026-07-05T06:36:55.687271Z","signed_message":"canonical_sha256_bytes"},"source_id":"2307.16125","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8afc525728495fa344a5d305da599ba5a7dbc93c24e3e1e593c0ff52c8e59ea0","sha256:fc00873123b823ad76d09765c7108480619db9e3825469f3816817307e0f6eb7"],"state_sha256":"65b2729a1c544e5ae43ab4ac3cd309bce50f7d049ca4d1190fd114a1ab71d95d"}