{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:B5YYWRS3WEMNU7OTSAH5DBNQZU","short_pith_number":"pith:B5YYWRS3","schema_version":"1.0","canonical_sha256":"0f718b465bb118da7dd3900fd185b0cd0c1a4b1123a3356746b8a665ab652176","source":{"kind":"arxiv","id":"2508.20478","version":2},"attestation_state":"computed","paper":{"title":"Video-MTR: Reinforced Multi-Turn Reasoning for Long Video Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lionel Ni, Tianshui Chen, Yuan Xie, Zheng Ge","submitted_at":"2025-08-28T06:55:08Z","abstract_excerpt":"Long-form video understanding, characterized by long-range temporal dependencies and multiple events, remains a challenge. Existing methods often rely on static reasoning or external visual-language models (VLMs), which face issues like complexity and sub-optimal performance due to the lack of end-to-end training. In this paper, we propose Video-MTR, a reinforced multi-turn reasoning framework designed to enable iterative key video segment selection and question comprehension. Unlike traditional video reasoning pipeline, which generate predictions in a single turn, Video-MTR performs reasoning"},"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":"2508.20478","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-28T06:55:08Z","cross_cats_sorted":[],"title_canon_sha256":"d73421037ce532a25edcc45ee081943e423151efaf70e81d4565cd501f116572","abstract_canon_sha256":"8d177dd486ce7ac8e26b5954b5b652fbcb4d6c0d932b53fcde284ea2da4d8dee"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:03:44.509182Z","signature_b64":"xQIKDz4h/IaP9x8cKKf3U0d/MZmpC+0fPZ5DwToWq8tbziloXI87GnInwxeqYF9aLuYGox7MF8f51FnuDiwiDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0f718b465bb118da7dd3900fd185b0cd0c1a4b1123a3356746b8a665ab652176","last_reissued_at":"2026-06-01T01:03:44.508043Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:03:44.508043Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Video-MTR: Reinforced Multi-Turn Reasoning for Long Video Understanding","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Lionel Ni, Tianshui Chen, Yuan Xie, Zheng Ge","submitted_at":"2025-08-28T06:55:08Z","abstract_excerpt":"Long-form video understanding, characterized by long-range temporal dependencies and multiple events, remains a challenge. Existing methods often rely on static reasoning or external visual-language models (VLMs), which face issues like complexity and sub-optimal performance due to the lack of end-to-end training. In this paper, we propose Video-MTR, a reinforced multi-turn reasoning framework designed to enable iterative key video segment selection and question comprehension. Unlike traditional video reasoning pipeline, which generate predictions in a single turn, Video-MTR performs reasoning"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.20478","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/2508.20478/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":"2508.20478","created_at":"2026-06-01T01:03:44.508212+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.20478v2","created_at":"2026-06-01T01:03:44.508212+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.20478","created_at":"2026-06-01T01:03:44.508212+00:00"},{"alias_kind":"pith_short_12","alias_value":"B5YYWRS3WEMN","created_at":"2026-06-01T01:03:44.508212+00:00"},{"alias_kind":"pith_short_16","alias_value":"B5YYWRS3WEMNU7OT","created_at":"2026-06-01T01:03:44.508212+00:00"},{"alias_kind":"pith_short_8","alias_value":"B5YYWRS3","created_at":"2026-06-01T01:03:44.508212+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":2,"sample":[{"citing_arxiv_id":"2511.13026","citing_title":"REVISOR: Beyond Textual Reflection, Towards Multimodal Introspective Reasoning in Long-Form Video Understanding","ref_index":51,"is_internal_anchor":true},{"citing_arxiv_id":"2601.15724","citing_title":"VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning","ref_index":37,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/B5YYWRS3WEMNU7OTSAH5DBNQZU","json":"https://pith.science/pith/B5YYWRS3WEMNU7OTSAH5DBNQZU.json","graph_json":"https://pith.science/api/pith-number/B5YYWRS3WEMNU7OTSAH5DBNQZU/graph.json","events_json":"https://pith.science/api/pith-number/B5YYWRS3WEMNU7OTSAH5DBNQZU/events.json","paper":"https://pith.science/paper/B5YYWRS3"},"agent_actions":{"view_html":"https://pith.science/pith/B5YYWRS3WEMNU7OTSAH5DBNQZU","download_json":"https://pith.science/pith/B5YYWRS3WEMNU7OTSAH5DBNQZU.json","view_paper":"https://pith.science/paper/B5YYWRS3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.20478&json=true","fetch_graph":"https://pith.science/api/pith-number/B5YYWRS3WEMNU7OTSAH5DBNQZU/graph.json","fetch_events":"https://pith.science/api/pith-number/B5YYWRS3WEMNU7OTSAH5DBNQZU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/B5YYWRS3WEMNU7OTSAH5DBNQZU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/B5YYWRS3WEMNU7OTSAH5DBNQZU/action/storage_attestation","attest_author":"https://pith.science/pith/B5YYWRS3WEMNU7OTSAH5DBNQZU/action/author_attestation","sign_citation":"https://pith.science/pith/B5YYWRS3WEMNU7OTSAH5DBNQZU/action/citation_signature","submit_replication":"https://pith.science/pith/B5YYWRS3WEMNU7OTSAH5DBNQZU/action/replication_record"}},"created_at":"2026-06-01T01:03:44.508212+00:00","updated_at":"2026-06-01T01:03:44.508212+00:00"}