{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:PF3DCNQYD3LZJ23PPY2F5S7M67","short_pith_number":"pith:PF3DCNQY","schema_version":"1.0","canonical_sha256":"79763136181ed794eb6f7e345ecbecf7ee8301cf4fbbf3fd750e4448fe0dd9b6","source":{"kind":"arxiv","id":"2503.13956","version":2},"attestation_state":"computed","paper":{"title":"Improving LLM Video Understanding with 16 Frames Per Second","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Changli Tang, Chao Zhang, Guangzhi Sun, Jimin Zhuang, Wei Li, Yixuan Li, Yudong Yang, Zejun Ma","submitted_at":"2025-03-18T06:48:08Z","abstract_excerpt":"Human vision is dynamic and continuous. However, in video understanding with multimodal large language models (LLMs), existing methods primarily rely on static features extracted from images sampled at a fixed low frame rate of frame-per-second (FPS) $\\leqslant$2, leading to critical visual information loss. In this paper, we introduce F-16, the first multimodal LLM designed for high-frame-rate video understanding. By increasing the frame rate to 16 FPS and compressing visual tokens within each 1-second clip, F-16 efficiently captures dynamic visual features while preserving key semantic infor"},"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":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2503.13956","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-03-18T06:48:08Z","cross_cats_sorted":[],"title_canon_sha256":"35b33fa16581cf6f16f4705f816edf1df2971b94da8efb503bab48abac0bf4f7","abstract_canon_sha256":"8473ea659c3028126ee4b21c8956c2e8f2dbec4bb52db753cf1ac6c85d153548"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:23:24.898662Z","signature_b64":"rm6Om3MWYHksb/phxzQ+wTGPlyoPdKzhKoLHNbvXY09MsKzkf0UZ2l4wU5dxuB08sfsP3N1a4fXdFLN5gkZfBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"79763136181ed794eb6f7e345ecbecf7ee8301cf4fbbf3fd750e4448fe0dd9b6","last_reissued_at":"2026-07-05T11:23:24.898026Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:23:24.898026Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving LLM Video Understanding with 16 Frames Per Second","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Changli Tang, Chao Zhang, Guangzhi Sun, Jimin Zhuang, Wei Li, Yixuan Li, Yudong Yang, Zejun Ma","submitted_at":"2025-03-18T06:48:08Z","abstract_excerpt":"Human vision is dynamic and continuous. However, in video understanding with multimodal large language models (LLMs), existing methods primarily rely on static features extracted from images sampled at a fixed low frame rate of frame-per-second (FPS) $\\leqslant$2, leading to critical visual information loss. In this paper, we introduce F-16, the first multimodal LLM designed for high-frame-rate video understanding. By increasing the frame rate to 16 FPS and compressing visual tokens within each 1-second clip, F-16 efficiently captures dynamic visual features while preserving key semantic infor"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.13956","kind":"arxiv","version":2},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2503.13956/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"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":"2503.13956","created_at":"2026-07-05T11:23:24.898090+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.13956v2","created_at":"2026-07-05T11:23:24.898090+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.13956","created_at":"2026-07-05T11:23:24.898090+00:00"},{"alias_kind":"pith_short_12","alias_value":"PF3DCNQYD3LZ","created_at":"2026-07-05T11:23:24.898090+00:00"},{"alias_kind":"pith_short_16","alias_value":"PF3DCNQYD3LZJ23P","created_at":"2026-07-05T11:23:24.898090+00:00"},{"alias_kind":"pith_short_8","alias_value":"PF3DCNQY","created_at":"2026-07-05T11:23:24.898090+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.20785","citing_title":"LongVT: Incentivizing \"Thinking with Long Videos\" via Native Tool Calling","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2604.08457","citing_title":"CrashSight: A Phase-Aware, Infrastructure-Centric Video Benchmark for Traffic Crash Scene Understanding and Reasoning","ref_index":13,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/PF3DCNQYD3LZJ23PPY2F5S7M67","json":"https://pith.science/pith/PF3DCNQYD3LZJ23PPY2F5S7M67.json","graph_json":"https://pith.science/api/pith-number/PF3DCNQYD3LZJ23PPY2F5S7M67/graph.json","events_json":"https://pith.science/api/pith-number/PF3DCNQYD3LZJ23PPY2F5S7M67/events.json","paper":"https://pith.science/paper/PF3DCNQY"},"agent_actions":{"view_html":"https://pith.science/pith/PF3DCNQYD3LZJ23PPY2F5S7M67","download_json":"https://pith.science/pith/PF3DCNQYD3LZJ23PPY2F5S7M67.json","view_paper":"https://pith.science/paper/PF3DCNQY","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.13956&json=true","fetch_graph":"https://pith.science/api/pith-number/PF3DCNQYD3LZJ23PPY2F5S7M67/graph.json","fetch_events":"https://pith.science/api/pith-number/PF3DCNQYD3LZJ23PPY2F5S7M67/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/PF3DCNQYD3LZJ23PPY2F5S7M67/action/timestamp_anchor","attest_storage":"https://pith.science/pith/PF3DCNQYD3LZJ23PPY2F5S7M67/action/storage_attestation","attest_author":"https://pith.science/pith/PF3DCNQYD3LZJ23PPY2F5S7M67/action/author_attestation","sign_citation":"https://pith.science/pith/PF3DCNQYD3LZJ23PPY2F5S7M67/action/citation_signature","submit_replication":"https://pith.science/pith/PF3DCNQYD3LZJ23PPY2F5S7M67/action/replication_record"}},"created_at":"2026-07-05T11:23:24.898090+00:00","updated_at":"2026-07-05T11:23:24.898090+00:00"}