{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:LWZ6PLF3WNKIGOBDQ4BCWAAS5N","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"88ed1e84f3bcb58687cf04f4079fe1c179668539b4cb1409199970239e037529","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-08-05T06:32:25Z","title_canon_sha256":"abb1979733dbe48e6c9322cc58ac4bc44c2c31973f6f553ede9613e4635e6803"},"schema_version":"1.0","source":{"id":"2608.04496","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.04496","created_at":"2026-08-06T01:35:42Z"},{"alias_kind":"arxiv_version","alias_value":"2608.04496v1","created_at":"2026-08-06T01:35:42Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.04496","created_at":"2026-08-06T01:35:42Z"},{"alias_kind":"pith_short_12","alias_value":"LWZ6PLF3WNKI","created_at":"2026-08-06T01:35:42Z"},{"alias_kind":"pith_short_16","alias_value":"LWZ6PLF3WNKIGOBD","created_at":"2026-08-06T01:35:42Z"},{"alias_kind":"pith_short_8","alias_value":"LWZ6PLF3","created_at":"2026-08-06T01:35:42Z"}],"graph_snapshots":[{"event_id":"sha256:bc5f5e9f57980c1db4a0c055fda157e47d451ef5bd72af59e2d8992a505c1ab2","target":"graph","created_at":"2026-08-06T01:35:42Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2608.04496/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Visual inputs in vision-language models (VLMs) are often encoded into substantially longer token sequences than text, making visual tokens a major bottleneck for efficient inference. Abundant recent methods address this bottleneck by scoring token importance and pruning low-scoring tokens in a single pass. However, one-shot scoring is insufficient because a token's prompt-relevant usefulness depends on the evidence already retained. Motivated by this insight, we introduce DIVE (Dynamic Iterative Visual Evidence Construction), a training-free framework that recasts visual-token pruning as dynam","authors_text":"Chen Zhong, Guangyi Yang, Jiepan Li, Wei He, Xiao An, Zijie Wang","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-08-05T06:32:25Z","title":"DIVE: Dynamic Iterative Visual Evidence Construction for Efficient Vision-Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.04496","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:0b93fd1bb987ba2cb89c9d6867d680807585e41e343ebdfc8d42597b34becd25","target":"record","created_at":"2026-08-06T01:35:42Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"88ed1e84f3bcb58687cf04f4079fe1c179668539b4cb1409199970239e037529","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-08-05T06:32:25Z","title_canon_sha256":"abb1979733dbe48e6c9322cc58ac4bc44c2c31973f6f553ede9613e4635e6803"},"schema_version":"1.0","source":{"id":"2608.04496","kind":"arxiv","version":1}},"canonical_sha256":"5db3e7acbbb35483382387022b0012eb5be66807d8a47934c2641c17e0b7b0de","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"5db3e7acbbb35483382387022b0012eb5be66807d8a47934c2641c17e0b7b0de","first_computed_at":"2026-08-06T01:35:42.857456Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-06T01:35:42.857456Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uEIHHnUUKnA/ANj4RsXNFxWXPywLED1soBoo5pq8IlHFFn3awBblbRdcBMsVhYEdGK33yEJr96qcXhhhfvhXDw==","signature_status":"signed_v1","signed_at":"2026-08-06T01:35:42.858835Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.04496","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:0b93fd1bb987ba2cb89c9d6867d680807585e41e343ebdfc8d42597b34becd25","sha256:bc5f5e9f57980c1db4a0c055fda157e47d451ef5bd72af59e2d8992a505c1ab2"],"state_sha256":"b78d295d52610609535cce97768b8e63c2bcb3878731f9dc53c02f3c29713770"}