{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:L55NJMBZE7TWUWJ6OL4K5UVVJO","short_pith_number":"pith:L55NJMBZ","schema_version":"1.0","canonical_sha256":"5f7ad4b03927e76a593e72f8aed2b54baafea2a6f4c64e9cdeb480586955ebd7","source":{"kind":"arxiv","id":"2501.08549","version":1},"attestation_state":"computed","paper":{"title":"The Devil is in Temporal Token: High Quality Video Reasoning Segmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Huchuan Lu, Lu Zhang, Pingping Zhang, Sitong Gong, Yunzhi Zhuge, Zongxin Yang","submitted_at":"2025-01-15T03:17:24Z","abstract_excerpt":"Existing methods for Video Reasoning Segmentation rely heavily on a single special token to represent the object in the keyframe or the entire video, inadequately capturing spatial complexity and inter-frame motion. To overcome these challenges, we propose VRS-HQ, an end-to-end video reasoning segmentation approach that leverages Multimodal Large Language Models (MLLMs) to inject rich spatiotemporal features into hierarchical tokens.Our key innovations include a Temporal Dynamic Aggregation (TDA) and a Token-driven Keyframe Selection (TKS). Specifically, we design frame-level <SEG> and tempora"},"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":"2501.08549","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-01-15T03:17:24Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"8e1a0adb0262b11059d350514a4ea41f32672aede1e885b27fe4001344014d0f","abstract_canon_sha256":"82bf3e719219325c43f009caf46f9e29b7245059ab35e6dc28b4f03526b09a23"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:21:28.140119Z","signature_b64":"dVmlAWzsqh+XDVeysQwFnDx4pNOjJY4FN6WkdtDrD2dRL8YV7hCVchtYT58x03QCdEZdG3TYP0ZhwAkPHBa3Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5f7ad4b03927e76a593e72f8aed2b54baafea2a6f4c64e9cdeb480586955ebd7","last_reissued_at":"2026-07-05T10:21:28.139534Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:21:28.139534Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Devil is in Temporal Token: High Quality Video Reasoning Segmentation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Huchuan Lu, Lu Zhang, Pingping Zhang, Sitong Gong, Yunzhi Zhuge, Zongxin Yang","submitted_at":"2025-01-15T03:17:24Z","abstract_excerpt":"Existing methods for Video Reasoning Segmentation rely heavily on a single special token to represent the object in the keyframe or the entire video, inadequately capturing spatial complexity and inter-frame motion. To overcome these challenges, we propose VRS-HQ, an end-to-end video reasoning segmentation approach that leverages Multimodal Large Language Models (MLLMs) to inject rich spatiotemporal features into hierarchical tokens.Our key innovations include a Temporal Dynamic Aggregation (TDA) and a Token-driven Keyframe Selection (TKS). Specifically, we design frame-level <SEG> and tempora"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.08549","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/2501.08549/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":"2501.08549","created_at":"2026-07-05T10:21:28.139598+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.08549v1","created_at":"2026-07-05T10:21:28.139598+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.08549","created_at":"2026-07-05T10:21:28.139598+00:00"},{"alias_kind":"pith_short_12","alias_value":"L55NJMBZE7TW","created_at":"2026-07-05T10:21:28.139598+00:00"},{"alias_kind":"pith_short_16","alias_value":"L55NJMBZE7TWUWJ6","created_at":"2026-07-05T10:21:28.139598+00:00"},{"alias_kind":"pith_short_8","alias_value":"L55NJMBZ","created_at":"2026-07-05T10:21:28.139598+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26196","citing_title":"From Structure to Synergy: A Survey of Vision-Language Perception Paradigm Evolution in Multimodal Large Language Models","ref_index":81,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/L55NJMBZE7TWUWJ6OL4K5UVVJO","json":"https://pith.science/pith/L55NJMBZE7TWUWJ6OL4K5UVVJO.json","graph_json":"https://pith.science/api/pith-number/L55NJMBZE7TWUWJ6OL4K5UVVJO/graph.json","events_json":"https://pith.science/api/pith-number/L55NJMBZE7TWUWJ6OL4K5UVVJO/events.json","paper":"https://pith.science/paper/L55NJMBZ"},"agent_actions":{"view_html":"https://pith.science/pith/L55NJMBZE7TWUWJ6OL4K5UVVJO","download_json":"https://pith.science/pith/L55NJMBZE7TWUWJ6OL4K5UVVJO.json","view_paper":"https://pith.science/paper/L55NJMBZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.08549&json=true","fetch_graph":"https://pith.science/api/pith-number/L55NJMBZE7TWUWJ6OL4K5UVVJO/graph.json","fetch_events":"https://pith.science/api/pith-number/L55NJMBZE7TWUWJ6OL4K5UVVJO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/L55NJMBZE7TWUWJ6OL4K5UVVJO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/L55NJMBZE7TWUWJ6OL4K5UVVJO/action/storage_attestation","attest_author":"https://pith.science/pith/L55NJMBZE7TWUWJ6OL4K5UVVJO/action/author_attestation","sign_citation":"https://pith.science/pith/L55NJMBZE7TWUWJ6OL4K5UVVJO/action/citation_signature","submit_replication":"https://pith.science/pith/L55NJMBZE7TWUWJ6OL4K5UVVJO/action/replication_record"}},"created_at":"2026-07-05T10:21:28.139598+00:00","updated_at":"2026-07-05T10:21:28.139598+00:00"}