{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:T5BHFWNKPFCLEXCO6O3MF4L24T","short_pith_number":"pith:T5BHFWNK","schema_version":"1.0","canonical_sha256":"9f4272d9aa7944b25c4ef3b6c2f17ae4dcbc818dbb473d251fa3c11849b3a7cf","source":{"kind":"arxiv","id":"2508.11256","version":1},"attestation_state":"computed","paper":{"title":"Generalized Decoupled Learning for Enhancing Open-Vocabulary Dense Perception","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bin Chen, Hengshuang Zhao, Junjie Wang, Keyu Chen, Xiaojuan Qi, Yulin Li, Zhuotao Tian","submitted_at":"2025-08-15T06:43:51Z","abstract_excerpt":"Dense visual perception tasks have been constrained by their reliance on predefined categories, limiting their applicability in real-world scenarios where visual concepts are unbounded. While Vision-Language Models (VLMs) like CLIP have shown promise in open-vocabulary tasks, their direct application to dense perception often leads to suboptimal performance due to limitations in local feature representation. In this work, we present our observation that CLIP's image tokens struggle to effectively aggregate information from spatially or semantically related regions, resulting in features that l"},"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":"2508.11256","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2025-08-15T06:43:51Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4c71fa94cdfbebf87b6d0b110712782887a4676e54616cda7db7fa538e3511d2","abstract_canon_sha256":"85ffb3d6dbc1205f3d02570696b1b002052471d376287784d4007b8dccd1fdbf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:54:21.535798Z","signature_b64":"QF4TbqJj9eARIR/GOF1CTnbAI9KaX1XZSFSZ++Sv2rPuGaDQpJLYwQjEiFQutXq10xQvSb18jvXjqgJQSK9TAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9f4272d9aa7944b25c4ef3b6c2f17ae4dcbc818dbb473d251fa3c11849b3a7cf","last_reissued_at":"2026-07-05T11:54:21.535435Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:54:21.535435Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generalized Decoupled Learning for Enhancing Open-Vocabulary Dense Perception","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Bin Chen, Hengshuang Zhao, Junjie Wang, Keyu Chen, Xiaojuan Qi, Yulin Li, Zhuotao Tian","submitted_at":"2025-08-15T06:43:51Z","abstract_excerpt":"Dense visual perception tasks have been constrained by their reliance on predefined categories, limiting their applicability in real-world scenarios where visual concepts are unbounded. While Vision-Language Models (VLMs) like CLIP have shown promise in open-vocabulary tasks, their direct application to dense perception often leads to suboptimal performance due to limitations in local feature representation. In this work, we present our observation that CLIP's image tokens struggle to effectively aggregate information from spatially or semantically related regions, resulting in features that l"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.11256","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/2508.11256/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":"2508.11256","created_at":"2026-07-05T11:54:21.535491+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.11256v1","created_at":"2026-07-05T11:54:21.535491+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.11256","created_at":"2026-07-05T11:54:21.535491+00:00"},{"alias_kind":"pith_short_12","alias_value":"T5BHFWNKPFCL","created_at":"2026-07-05T11:54:21.535491+00:00"},{"alias_kind":"pith_short_16","alias_value":"T5BHFWNKPFCLEXCO","created_at":"2026-07-05T11:54:21.535491+00:00"},{"alias_kind":"pith_short_8","alias_value":"T5BHFWNK","created_at":"2026-07-05T11:54:21.535491+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.26120","citing_title":"Dynamic-dLLM: Dynamic Cache-Budget and Adaptive Parallel Decoding for Training-Free Acceleration of Diffusion LLM","ref_index":14,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17583","citing_title":"AgentSteerTTS: A Multi-Agent Closed-Loop Framework for Composite-Instruction Text-to-Speech","ref_index":105,"is_internal_anchor":false},{"citing_arxiv_id":"2604.17914","citing_title":"Beyond Binary Contrast: Modeling Continuous Skeleton Action Spaces with Transitional Anchors","ref_index":60,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/T5BHFWNKPFCLEXCO6O3MF4L24T","json":"https://pith.science/pith/T5BHFWNKPFCLEXCO6O3MF4L24T.json","graph_json":"https://pith.science/api/pith-number/T5BHFWNKPFCLEXCO6O3MF4L24T/graph.json","events_json":"https://pith.science/api/pith-number/T5BHFWNKPFCLEXCO6O3MF4L24T/events.json","paper":"https://pith.science/paper/T5BHFWNK"},"agent_actions":{"view_html":"https://pith.science/pith/T5BHFWNKPFCLEXCO6O3MF4L24T","download_json":"https://pith.science/pith/T5BHFWNKPFCLEXCO6O3MF4L24T.json","view_paper":"https://pith.science/paper/T5BHFWNK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.11256&json=true","fetch_graph":"https://pith.science/api/pith-number/T5BHFWNKPFCLEXCO6O3MF4L24T/graph.json","fetch_events":"https://pith.science/api/pith-number/T5BHFWNKPFCLEXCO6O3MF4L24T/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/T5BHFWNKPFCLEXCO6O3MF4L24T/action/timestamp_anchor","attest_storage":"https://pith.science/pith/T5BHFWNKPFCLEXCO6O3MF4L24T/action/storage_attestation","attest_author":"https://pith.science/pith/T5BHFWNKPFCLEXCO6O3MF4L24T/action/author_attestation","sign_citation":"https://pith.science/pith/T5BHFWNKPFCLEXCO6O3MF4L24T/action/citation_signature","submit_replication":"https://pith.science/pith/T5BHFWNKPFCLEXCO6O3MF4L24T/action/replication_record"}},"created_at":"2026-07-05T11:54:21.535491+00:00","updated_at":"2026-07-05T11:54:21.535491+00:00"}