{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:SRURN5W6VVIDUXD3HUJS2QHF76","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":"cc94e71e283553cc5c82f87cfaaa94def5c896a7d0dacc4197f2fbcfc7c23609","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2023-11-20T16:02:10Z","title_canon_sha256":"030d90bc51886da6dffed60e26e226f848cbd88713b3b266523407798dffe0ad"},"schema_version":"1.0","source":{"id":"2311.11865","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2311.11865","created_at":"2026-07-05T07:14:40Z"},{"alias_kind":"arxiv_version","alias_value":"2311.11865v1","created_at":"2026-07-05T07:14:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.11865","created_at":"2026-07-05T07:14:40Z"},{"alias_kind":"pith_short_12","alias_value":"SRURN5W6VVID","created_at":"2026-07-05T07:14:40Z"},{"alias_kind":"pith_short_16","alias_value":"SRURN5W6VVIDUXD3","created_at":"2026-07-05T07:14:40Z"},{"alias_kind":"pith_short_8","alias_value":"SRURN5W6","created_at":"2026-07-05T07:14:40Z"}],"graph_snapshots":[{"event_id":"sha256:2e3cee896502a14da4592d3f5b67d3274afa2cb91e6144a0f9334c8a17299e7c","target":"graph","created_at":"2026-07-05T07:14:40Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2311.11865/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Despite the rapid development of video Large Language Models (LLMs), a comprehensive evaluation is still absent. In this paper, we introduce a unified evaluation that encompasses multiple video tasks, including captioning, question and answering, retrieval, and action recognition. In addition to conventional metrics, we showcase how GPT-based evaluation can match human-like performance in assessing response quality across multiple aspects. We propose a simple baseline: Video-LLaVA, which uses a single linear projection and outperforms existing video LLMs. Finally, we evaluate video LLMs beyond","authors_text":"Fan Jia, Qiuyue Wang, Shuailin Li, Tiancai Wang, Yingfei Liu, Yuang Zhang, Yucheng Zhao","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2023-11-20T16:02:10Z","title":"VLM-Eval: A General Evaluation on Video Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.11865","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:92ca11bd94ddf249d644694e3be46a829579c596736d6b91aee821c17fc73798","target":"record","created_at":"2026-07-05T07:14:40Z","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":"cc94e71e283553cc5c82f87cfaaa94def5c896a7d0dacc4197f2fbcfc7c23609","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2023-11-20T16:02:10Z","title_canon_sha256":"030d90bc51886da6dffed60e26e226f848cbd88713b3b266523407798dffe0ad"},"schema_version":"1.0","source":{"id":"2311.11865","kind":"arxiv","version":1}},"canonical_sha256":"946916f6dead503a5c7b3d132d40e5ffad51d2a23a1a2b59cf26686ebbcb3314","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"946916f6dead503a5c7b3d132d40e5ffad51d2a23a1a2b59cf26686ebbcb3314","first_computed_at":"2026-07-05T07:14:40.127172Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T07:14:40.127172Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"S5tk1esorv4s7b+iAax0s5AJQ9rePXsK/HnKR2HEzuGG2NVVz9kv28NKM44mevWbgHiKl9gzVQObq2SvXO3CDg==","signature_status":"signed_v1","signed_at":"2026-07-05T07:14:40.127643Z","signed_message":"canonical_sha256_bytes"},"source_id":"2311.11865","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:92ca11bd94ddf249d644694e3be46a829579c596736d6b91aee821c17fc73798","sha256:2e3cee896502a14da4592d3f5b67d3274afa2cb91e6144a0f9334c8a17299e7c"],"state_sha256":"d0730c8ea68b0ca776d88e8a796127a54df47b0310baa048b33064aa3ff15638"}