{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:KAZASH4JDLOHH75TZ7TQ2MGEXQ","short_pith_number":"pith:KAZASH4J","schema_version":"1.0","canonical_sha256":"5032091f891adc73ffb3cfe70d30c4bc04b48e20783fb5d2023bfd7ece529235","source":{"kind":"arxiv","id":"2606.18439","version":1},"attestation_state":"computed","paper":{"title":"RegimeVGGT: Layer-Wise Spatially Preserving Redundancy Removal for Visual Geometry Grounded Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.CV","authors_text":"(2) University of California, (3) Nanyang Technological University), Irvine, Jiaxiang Hu (2), Jinhao You (1), Kai Tang (3), Shuo Lyu (1), Tanxuan Li (1), Yichen Guo (3) ((1) University of Pennsylvania, Zhuohang Lyu (1), Zibo Zhao (1)","submitted_at":"2026-06-16T19:41:23Z","abstract_excerpt":"Visual Geometry Grounded Transformer (VGGT) recovers dense 3D scene structure from multi-view images in one forward pass, but quadratic cross-frame attention limits its scalability. Existing training-free accelerators reduce computation uniformly along one axis, missing layer heterogeneity. Our spectral, probing, and causal analyses reveal three regimes: shallow layers lack cross-view structure, middle layers drive cross-view alignment, and deep layers are redundant for dense geometry yet their cross-frame attention remains essential for pose. RegimeVGGT applies layer-wise U-shaped compression"},"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":"2606.18439","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-16T19:41:23Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"414939b9a8bb43657c7be7687e89fc2b023b6c8441a1204019df931a57e6ad8a","abstract_canon_sha256":"8ea388cf755e6ba2413de43513b12ac0367dda333d95ab318c23a7b7c8a7e263"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-19T16:11:01.237429Z","signature_b64":"uA0MQEDdYjO3QepgdFD3TCvnWJz0rYh6WGHCNkXj0NHkTt4YDBPkXFBoOHmsDshqHqws5KSuGWl4ruvr/sWJCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5032091f891adc73ffb3cfe70d30c4bc04b48e20783fb5d2023bfd7ece529235","last_reissued_at":"2026-06-19T16:11:01.237002Z","signature_status":"signed_v1","first_computed_at":"2026-06-19T16:11:01.237002Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RegimeVGGT: Layer-Wise Spatially Preserving Redundancy Removal for Visual Geometry Grounded Transformer","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.CV","authors_text":"(2) University of California, (3) Nanyang Technological University), Irvine, Jiaxiang Hu (2), Jinhao You (1), Kai Tang (3), Shuo Lyu (1), Tanxuan Li (1), Yichen Guo (3) ((1) University of Pennsylvania, Zhuohang Lyu (1), Zibo Zhao (1)","submitted_at":"2026-06-16T19:41:23Z","abstract_excerpt":"Visual Geometry Grounded Transformer (VGGT) recovers dense 3D scene structure from multi-view images in one forward pass, but quadratic cross-frame attention limits its scalability. Existing training-free accelerators reduce computation uniformly along one axis, missing layer heterogeneity. Our spectral, probing, and causal analyses reveal three regimes: shallow layers lack cross-view structure, middle layers drive cross-view alignment, and deep layers are redundant for dense geometry yet their cross-frame attention remains essential for pose. RegimeVGGT applies layer-wise U-shaped compression"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.18439","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/2606.18439/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":"2606.18439","created_at":"2026-06-19T16:11:01.237071+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.18439v1","created_at":"2026-06-19T16:11:01.237071+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.18439","created_at":"2026-06-19T16:11:01.237071+00:00"},{"alias_kind":"pith_short_12","alias_value":"KAZASH4JDLOH","created_at":"2026-06-19T16:11:01.237071+00:00"},{"alias_kind":"pith_short_16","alias_value":"KAZASH4JDLOHH75T","created_at":"2026-06-19T16:11:01.237071+00:00"},{"alias_kind":"pith_short_8","alias_value":"KAZASH4J","created_at":"2026-06-19T16:11:01.237071+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/KAZASH4JDLOHH75TZ7TQ2MGEXQ","json":"https://pith.science/pith/KAZASH4JDLOHH75TZ7TQ2MGEXQ.json","graph_json":"https://pith.science/api/pith-number/KAZASH4JDLOHH75TZ7TQ2MGEXQ/graph.json","events_json":"https://pith.science/api/pith-number/KAZASH4JDLOHH75TZ7TQ2MGEXQ/events.json","paper":"https://pith.science/paper/KAZASH4J"},"agent_actions":{"view_html":"https://pith.science/pith/KAZASH4JDLOHH75TZ7TQ2MGEXQ","download_json":"https://pith.science/pith/KAZASH4JDLOHH75TZ7TQ2MGEXQ.json","view_paper":"https://pith.science/paper/KAZASH4J","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.18439&json=true","fetch_graph":"https://pith.science/api/pith-number/KAZASH4JDLOHH75TZ7TQ2MGEXQ/graph.json","fetch_events":"https://pith.science/api/pith-number/KAZASH4JDLOHH75TZ7TQ2MGEXQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KAZASH4JDLOHH75TZ7TQ2MGEXQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KAZASH4JDLOHH75TZ7TQ2MGEXQ/action/storage_attestation","attest_author":"https://pith.science/pith/KAZASH4JDLOHH75TZ7TQ2MGEXQ/action/author_attestation","sign_citation":"https://pith.science/pith/KAZASH4JDLOHH75TZ7TQ2MGEXQ/action/citation_signature","submit_replication":"https://pith.science/pith/KAZASH4JDLOHH75TZ7TQ2MGEXQ/action/replication_record"}},"created_at":"2026-06-19T16:11:01.237071+00:00","updated_at":"2026-06-19T16:11:01.237071+00:00"}