{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:N6BNH32BFLZF2BF2RJU2H73SPT","short_pith_number":"pith:N6BNH32B","schema_version":"1.0","canonical_sha256":"6f82d3ef412af25d04ba8a69a3ff727cd348e508e64bbed0c2b75b372d3461ec","source":{"kind":"arxiv","id":"2404.00308","version":1},"attestation_state":"computed","paper":{"title":"ST-LLM: Large Language Models Are Effective Temporal Learners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Li, Ge Li, Haoran Tang, Ruyang Liu, Ying Shan, Yixiao Ge","submitted_at":"2024-03-30T10:11:26Z","abstract_excerpt":"Large Language Models (LLMs) have showcased impressive capabilities in text comprehension and generation, prompting research efforts towards video LLMs to facilitate human-AI interaction at the video level. However, how to effectively encode and understand videos in video-based dialogue systems remains to be solved. In this paper, we investigate a straightforward yet unexplored question: Can we feed all spatial-temporal tokens into the LLM, thus delegating the task of video sequence modeling to the LLMs? Surprisingly, this simple approach yields significant improvements in video understanding."},"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":"2404.00308","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2024-03-30T10:11:26Z","cross_cats_sorted":[],"title_canon_sha256":"646c7967b12ce376339442710584650e3742dc8487239502720d4b1aa439c423","abstract_canon_sha256":"46838f981b2c77726b9bb288889bd35392de84fcc860a71d733c917bb229b68d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:02:31.290743Z","signature_b64":"CmEfXoy6B8pLm1at51thDXpa6c+ay3xxoe4kD/Uth6vin/7Pfi5rTQS9O9aBjvnyFfgFeHIDGbC9CyBvjiVxAg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6f82d3ef412af25d04ba8a69a3ff727cd348e508e64bbed0c2b75b372d3461ec","last_reissued_at":"2026-07-05T08:02:31.290213Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:02:31.290213Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ST-LLM: Large Language Models Are Effective Temporal Learners","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chen Li, Ge Li, Haoran Tang, Ruyang Liu, Ying Shan, Yixiao Ge","submitted_at":"2024-03-30T10:11:26Z","abstract_excerpt":"Large Language Models (LLMs) have showcased impressive capabilities in text comprehension and generation, prompting research efforts towards video LLMs to facilitate human-AI interaction at the video level. However, how to effectively encode and understand videos in video-based dialogue systems remains to be solved. In this paper, we investigate a straightforward yet unexplored question: Can we feed all spatial-temporal tokens into the LLM, thus delegating the task of video sequence modeling to the LLMs? Surprisingly, this simple approach yields significant improvements in video understanding."},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2404.00308","kind":"arxiv","version":1},"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/2404.00308/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":"2404.00308","created_at":"2026-07-05T08:02:31.290283+00:00"},{"alias_kind":"arxiv_version","alias_value":"2404.00308v1","created_at":"2026-07-05T08:02:31.290283+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2404.00308","created_at":"2026-07-05T08:02:31.290283+00:00"},{"alias_kind":"pith_short_12","alias_value":"N6BNH32BFLZF","created_at":"2026-07-05T08:02:31.290283+00:00"},{"alias_kind":"pith_short_16","alias_value":"N6BNH32BFLZF2BF2","created_at":"2026-07-05T08:02:31.290283+00:00"},{"alias_kind":"pith_short_8","alias_value":"N6BNH32B","created_at":"2026-07-05T08:02:31.290283+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.19706","citing_title":"NEST: Narrative Event Structures in Time for Long Video Understanding","ref_index":284,"is_internal_anchor":false},{"citing_arxiv_id":"2501.05067","citing_title":"LLaVA-Octopus: Unlocking Instruction-Driven Adaptive Projector Fusion for Video Understanding","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2408.16500","citing_title":"CogVLM2: Visual Language Models for Image and Video Understanding","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2404.16994","citing_title":"PLLaVA : Parameter-free LLaVA Extension from Images to Videos for Video Dense Captioning","ref_index":31,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/N6BNH32BFLZF2BF2RJU2H73SPT","json":"https://pith.science/pith/N6BNH32BFLZF2BF2RJU2H73SPT.json","graph_json":"https://pith.science/api/pith-number/N6BNH32BFLZF2BF2RJU2H73SPT/graph.json","events_json":"https://pith.science/api/pith-number/N6BNH32BFLZF2BF2RJU2H73SPT/events.json","paper":"https://pith.science/paper/N6BNH32B"},"agent_actions":{"view_html":"https://pith.science/pith/N6BNH32BFLZF2BF2RJU2H73SPT","download_json":"https://pith.science/pith/N6BNH32BFLZF2BF2RJU2H73SPT.json","view_paper":"https://pith.science/paper/N6BNH32B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2404.00308&json=true","fetch_graph":"https://pith.science/api/pith-number/N6BNH32BFLZF2BF2RJU2H73SPT/graph.json","fetch_events":"https://pith.science/api/pith-number/N6BNH32BFLZF2BF2RJU2H73SPT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/N6BNH32BFLZF2BF2RJU2H73SPT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/N6BNH32BFLZF2BF2RJU2H73SPT/action/storage_attestation","attest_author":"https://pith.science/pith/N6BNH32BFLZF2BF2RJU2H73SPT/action/author_attestation","sign_citation":"https://pith.science/pith/N6BNH32BFLZF2BF2RJU2H73SPT/action/citation_signature","submit_replication":"https://pith.science/pith/N6BNH32BFLZF2BF2RJU2H73SPT/action/replication_record"}},"created_at":"2026-07-05T08:02:31.290283+00:00","updated_at":"2026-07-05T08:02:31.290283+00:00"}