{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:TCBOIWDF4J7LK3ZP6K7ZY3BGUL","short_pith_number":"pith:TCBOIWDF","schema_version":"1.0","canonical_sha256":"9882e45865e27eb56f2ff2bf9c6c26a2e16ac2035f20166f3b24b6261a8a3ffe","source":{"kind":"arxiv","id":"2608.01821","version":1},"attestation_state":"computed","paper":{"title":"DAVET: Denoising-Aware Visual Evidence Trajectory Allocation for Diffusion Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Cheng Yan, Fan Xu, Wuyang Zhang, Xiang Xia, Yongkang Zhou","submitted_at":"2026-08-03T07:29:05Z","abstract_excerpt":"Diffusion vision-language models (dVLMs) iteratively denoise masked responses while conditioning each denoising step on visual evidence, making visual conditioning a substantial recurring inference cost. Unlike autoregressive decoding, diffusion generation repeatedly revisits the entire response as uncertainty evolves. Our analysis reveals that visual evidence demand is strongly step-dependent, motivating adaptive allocation across denoising steps. Existing inference acceleration methods operate through decoding-side strategies or visual token compression via pruning and merging, but do not ex"},"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":"2608.01821","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-08-03T07:29:05Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"9e0df7fb11257a73b7bb9dc973aad22f07d59cb6b1a481ed87c02b1559cf3cc4","abstract_canon_sha256":"77f2683f813589d1196ccf01bf3fb2c5d59870144b83764fba9e19baab060678"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-04T02:07:14.631523Z","signature_b64":"7um+htX3ER2lasUMfwVAqXoalGncNShGh6JxbDPgDqhzWJstehJN5fs1hiuBJ8VB7qIn7VHnpxw2WBqej0fcAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9882e45865e27eb56f2ff2bf9c6c26a2e16ac2035f20166f3b24b6261a8a3ffe","last_reissued_at":"2026-08-04T02:07:14.629916Z","signature_status":"signed_v1","first_computed_at":"2026-08-04T02:07:14.629916Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DAVET: Denoising-Aware Visual Evidence Trajectory Allocation for Diffusion Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CV","authors_text":"Cheng Yan, Fan Xu, Wuyang Zhang, Xiang Xia, Yongkang Zhou","submitted_at":"2026-08-03T07:29:05Z","abstract_excerpt":"Diffusion vision-language models (dVLMs) iteratively denoise masked responses while conditioning each denoising step on visual evidence, making visual conditioning a substantial recurring inference cost. Unlike autoregressive decoding, diffusion generation repeatedly revisits the entire response as uncertainty evolves. Our analysis reveals that visual evidence demand is strongly step-dependent, motivating adaptive allocation across denoising steps. Existing inference acceleration methods operate through decoding-side strategies or visual token compression via pruning and merging, but do not ex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.01821","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/2608.01821/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":"2608.01821","created_at":"2026-08-04T02:07:14.631408+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.01821v1","created_at":"2026-08-04T02:07:14.631408+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.01821","created_at":"2026-08-04T02:07:14.631408+00:00"},{"alias_kind":"pith_short_12","alias_value":"TCBOIWDF4J7L","created_at":"2026-08-04T02:07:14.631408+00:00"},{"alias_kind":"pith_short_16","alias_value":"TCBOIWDF4J7LK3ZP","created_at":"2026-08-04T02:07:14.631408+00:00"},{"alias_kind":"pith_short_8","alias_value":"TCBOIWDF","created_at":"2026-08-04T02:07:14.631408+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/TCBOIWDF4J7LK3ZP6K7ZY3BGUL","json":"https://pith.science/pith/TCBOIWDF4J7LK3ZP6K7ZY3BGUL.json","graph_json":"https://pith.science/api/pith-number/TCBOIWDF4J7LK3ZP6K7ZY3BGUL/graph.json","events_json":"https://pith.science/api/pith-number/TCBOIWDF4J7LK3ZP6K7ZY3BGUL/events.json","paper":"https://pith.science/paper/TCBOIWDF"},"agent_actions":{"view_html":"https://pith.science/pith/TCBOIWDF4J7LK3ZP6K7ZY3BGUL","download_json":"https://pith.science/pith/TCBOIWDF4J7LK3ZP6K7ZY3BGUL.json","view_paper":"https://pith.science/paper/TCBOIWDF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.01821&json=true","fetch_graph":"https://pith.science/api/pith-number/TCBOIWDF4J7LK3ZP6K7ZY3BGUL/graph.json","fetch_events":"https://pith.science/api/pith-number/TCBOIWDF4J7LK3ZP6K7ZY3BGUL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TCBOIWDF4J7LK3ZP6K7ZY3BGUL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TCBOIWDF4J7LK3ZP6K7ZY3BGUL/action/storage_attestation","attest_author":"https://pith.science/pith/TCBOIWDF4J7LK3ZP6K7ZY3BGUL/action/author_attestation","sign_citation":"https://pith.science/pith/TCBOIWDF4J7LK3ZP6K7ZY3BGUL/action/citation_signature","submit_replication":"https://pith.science/pith/TCBOIWDF4J7LK3ZP6K7ZY3BGUL/action/replication_record"}},"created_at":"2026-08-04T02:07:14.631408+00:00","updated_at":"2026-08-04T02:07:14.631408+00:00"}