{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:AF7JJ3EHBYA5VLGRURRZ6MK4UT","short_pith_number":"pith:AF7JJ3EH","schema_version":"1.0","canonical_sha256":"017e94ec870e01daacd1a4639f315ca4ca0e2190bcee5834d029c9b570f528ae","source":{"kind":"arxiv","id":"2607.17423","version":1},"attestation_state":"computed","paper":{"title":"TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Changlian Ma, Haoran Chen, Jun Zhang, Limin Wang, Shaowei Zhang, Songze Li, Tianxiang Jiang, Xiangyu Zeng, Xinhao Li, Xinyu Chen, Yuandong Yang, Yuhan Zhu, Zhiqiu Zhang, Ziang Yan, Zikang Wang","submitted_at":"2026-07-19T22:04:33Z","abstract_excerpt":"Video multimodal large language models (MLLMs) can describe what happens in a video, but rarely identify when the supporting evidence occurs. We study generalist video temporal grounding, in which one model predicts a variable-cardinality set of evidence intervals across video lengths, domains, query forms, and viewpoints. Existing training strategies are misaligned with this set-valued task: long-video labels often rely on brittle one-pass annotation, while reinforcement-learning rewards either fail to distinguish non-overlapping predictions or require fragile segment matching. TimeLens2 trea"},"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":"2607.17423","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-19T22:04:33Z","cross_cats_sorted":[],"title_canon_sha256":"7a970514029f25581b0c50a202de32be3d1e4ebba9a882aca5d0f41dca0319ab","abstract_canon_sha256":"c4ce991312b0a46d4e3b0c6d81e75653943e9e6010f6fd22a4cd1130877f30a4"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-21T01:21:32.262689Z","signature_b64":"9UyA5CS0hp1ji+MSP6QoknAGiqBNSvLcmMRE9liTttp11XweZPk9SOGliocpsSJbb7H5svM2Rn8iJw1QVrUSAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"017e94ec870e01daacd1a4639f315ca4ca0e2190bcee5834d029c9b570f528ae","last_reissued_at":"2026-07-21T01:21:32.261815Z","signature_status":"signed_v1","first_computed_at":"2026-07-21T01:21:32.261815Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Changlian Ma, Haoran Chen, Jun Zhang, Limin Wang, Shaowei Zhang, Songze Li, Tianxiang Jiang, Xiangyu Zeng, Xinhao Li, Xinyu Chen, Yuandong Yang, Yuhan Zhu, Zhiqiu Zhang, Ziang Yan, Zikang Wang","submitted_at":"2026-07-19T22:04:33Z","abstract_excerpt":"Video multimodal large language models (MLLMs) can describe what happens in a video, but rarely identify when the supporting evidence occurs. We study generalist video temporal grounding, in which one model predicts a variable-cardinality set of evidence intervals across video lengths, domains, query forms, and viewpoints. Existing training strategies are misaligned with this set-valued task: long-video labels often rely on brittle one-pass annotation, while reinforcement-learning rewards either fail to distinguish non-overlapping predictions or require fragile segment matching. TimeLens2 trea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.17423","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/2607.17423/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":"2607.17423","created_at":"2026-07-21T01:21:32.262265+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.17423v1","created_at":"2026-07-21T01:21:32.262265+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.17423","created_at":"2026-07-21T01:21:32.262265+00:00"},{"alias_kind":"pith_short_12","alias_value":"AF7JJ3EHBYA5","created_at":"2026-07-21T01:21:32.262265+00:00"},{"alias_kind":"pith_short_16","alias_value":"AF7JJ3EHBYA5VLGR","created_at":"2026-07-21T01:21:32.262265+00:00"},{"alias_kind":"pith_short_8","alias_value":"AF7JJ3EH","created_at":"2026-07-21T01:21:32.262265+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/AF7JJ3EHBYA5VLGRURRZ6MK4UT","json":"https://pith.science/pith/AF7JJ3EHBYA5VLGRURRZ6MK4UT.json","graph_json":"https://pith.science/api/pith-number/AF7JJ3EHBYA5VLGRURRZ6MK4UT/graph.json","events_json":"https://pith.science/api/pith-number/AF7JJ3EHBYA5VLGRURRZ6MK4UT/events.json","paper":"https://pith.science/paper/AF7JJ3EH"},"agent_actions":{"view_html":"https://pith.science/pith/AF7JJ3EHBYA5VLGRURRZ6MK4UT","download_json":"https://pith.science/pith/AF7JJ3EHBYA5VLGRURRZ6MK4UT.json","view_paper":"https://pith.science/paper/AF7JJ3EH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.17423&json=true","fetch_graph":"https://pith.science/api/pith-number/AF7JJ3EHBYA5VLGRURRZ6MK4UT/graph.json","fetch_events":"https://pith.science/api/pith-number/AF7JJ3EHBYA5VLGRURRZ6MK4UT/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AF7JJ3EHBYA5VLGRURRZ6MK4UT/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AF7JJ3EHBYA5VLGRURRZ6MK4UT/action/storage_attestation","attest_author":"https://pith.science/pith/AF7JJ3EHBYA5VLGRURRZ6MK4UT/action/author_attestation","sign_citation":"https://pith.science/pith/AF7JJ3EHBYA5VLGRURRZ6MK4UT/action/citation_signature","submit_replication":"https://pith.science/pith/AF7JJ3EHBYA5VLGRURRZ6MK4UT/action/replication_record"}},"created_at":"2026-07-21T01:21:32.262265+00:00","updated_at":"2026-07-21T01:21:32.262265+00:00"}