{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:Z2QPWY6DHUTQEMO7YUMAOMG6QZ","short_pith_number":"pith:Z2QPWY6D","schema_version":"1.0","canonical_sha256":"cea0fb63c33d270231dfc5180730de864e7743fc6af2d53cbc5e31f32b9bbeb4","source":{"kind":"arxiv","id":"2606.01756","version":1},"attestation_state":"computed","paper":{"title":"EvoCut: Multi-Layer Evolution-Aware Visual Token Compression for Efficient Large Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Zhang, Hongyu Lu, Huanling Hu, Jiawei Li, Pengfei Zhang, Shikai Jiang, Wenwei Jin, Yao Hu","submitted_at":"2026-06-01T06:21:28Z","abstract_excerpt":"Large vision-language models (LVLMs) achieve strong performance on image and video understanding tasks, but their inference efficiency is constrained by the large number of visual tokens produced by vision encoders. Most existing visual token compression methods estimate token importance from attention scores or representation properties at specific layers, overlooking how visual tokens evolve across the vision encoder. Such layer-specific criteria may provide incomplete importance estimates and limit performance preservation after compression. To address this issue, we analyze layer-wise visu"},"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":"2606.01756","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-06-01T06:21:28Z","cross_cats_sorted":[],"title_canon_sha256":"c3097bbfc4fde8e2aa8d0dbf69aedd8df41eb935e8bd02b18605f2ffe074c6d5","abstract_canon_sha256":"0dc16b670d8310ad60e498a0cf171a7dd17fc30d7b841ad8bce33c7860d044c8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T02:04:55.844266Z","signature_b64":"JShZjYv6XPOpsTW8ZQ7E+TtrhHUblwfsgLYpx20AcsO5nfvQQqRgDtOuwT54DlHIfg8gBg4RdT6Otr53bNr9Aw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cea0fb63c33d270231dfc5180730de864e7743fc6af2d53cbc5e31f32b9bbeb4","last_reissued_at":"2026-06-02T02:04:55.843842Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T02:04:55.843842Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"EvoCut: Multi-Layer Evolution-Aware Visual Token Compression for Efficient Large Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Feng Zhang, Hongyu Lu, Huanling Hu, Jiawei Li, Pengfei Zhang, Shikai Jiang, Wenwei Jin, Yao Hu","submitted_at":"2026-06-01T06:21:28Z","abstract_excerpt":"Large vision-language models (LVLMs) achieve strong performance on image and video understanding tasks, but their inference efficiency is constrained by the large number of visual tokens produced by vision encoders. Most existing visual token compression methods estimate token importance from attention scores or representation properties at specific layers, overlooking how visual tokens evolve across the vision encoder. Such layer-specific criteria may provide incomplete importance estimates and limit performance preservation after compression. To address this issue, we analyze layer-wise visu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.01756","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/2606.01756/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":"2606.01756","created_at":"2026-06-02T02:04:55.843920+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.01756v1","created_at":"2026-06-02T02:04:55.843920+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.01756","created_at":"2026-06-02T02:04:55.843920+00:00"},{"alias_kind":"pith_short_12","alias_value":"Z2QPWY6DHUTQ","created_at":"2026-06-02T02:04:55.843920+00:00"},{"alias_kind":"pith_short_16","alias_value":"Z2QPWY6DHUTQEMO7","created_at":"2026-06-02T02:04:55.843920+00:00"},{"alias_kind":"pith_short_8","alias_value":"Z2QPWY6D","created_at":"2026-06-02T02:04:55.843920+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/Z2QPWY6DHUTQEMO7YUMAOMG6QZ","json":"https://pith.science/pith/Z2QPWY6DHUTQEMO7YUMAOMG6QZ.json","graph_json":"https://pith.science/api/pith-number/Z2QPWY6DHUTQEMO7YUMAOMG6QZ/graph.json","events_json":"https://pith.science/api/pith-number/Z2QPWY6DHUTQEMO7YUMAOMG6QZ/events.json","paper":"https://pith.science/paper/Z2QPWY6D"},"agent_actions":{"view_html":"https://pith.science/pith/Z2QPWY6DHUTQEMO7YUMAOMG6QZ","download_json":"https://pith.science/pith/Z2QPWY6DHUTQEMO7YUMAOMG6QZ.json","view_paper":"https://pith.science/paper/Z2QPWY6D","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.01756&json=true","fetch_graph":"https://pith.science/api/pith-number/Z2QPWY6DHUTQEMO7YUMAOMG6QZ/graph.json","fetch_events":"https://pith.science/api/pith-number/Z2QPWY6DHUTQEMO7YUMAOMG6QZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/Z2QPWY6DHUTQEMO7YUMAOMG6QZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/Z2QPWY6DHUTQEMO7YUMAOMG6QZ/action/storage_attestation","attest_author":"https://pith.science/pith/Z2QPWY6DHUTQEMO7YUMAOMG6QZ/action/author_attestation","sign_citation":"https://pith.science/pith/Z2QPWY6DHUTQEMO7YUMAOMG6QZ/action/citation_signature","submit_replication":"https://pith.science/pith/Z2QPWY6DHUTQEMO7YUMAOMG6QZ/action/replication_record"}},"created_at":"2026-06-02T02:04:55.843920+00:00","updated_at":"2026-06-02T02:04:55.843920+00:00"}