{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:CSB6DU6FXNLBXZDF3VY76WEMZW","short_pith_number":"pith:CSB6DU6F","schema_version":"1.0","canonical_sha256":"1483e1d3c5bb561be465dd71ff588ccd9c6ca4732998ecd8fb7229898de6027f","source":{"kind":"arxiv","id":"2607.23373","version":1},"attestation_state":"computed","paper":{"title":"UltraViT: Latency-Optimized On-device Vision Encoder for Large Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adrian Bulat, Alberto Baldrati, Anestis Zaganidis, Georgios Tzimiropoulos, Hyeonuk Kim, Ioannis Maniadis Metaxas, Yassine Ouali","submitted_at":"2026-07-25T21:31:40Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) remain bottlenecked by massive computational footprints, precluding their deployment on resource-constrained edge devices. While efforts to compress LVLMs focus heavily on vision token reduction or smaller language models, the vision encoder is largely overlooked, typically deployed as a monolithic, computationally heavy feature extractor. Moreover, there is no previous effort that designs a vision encoder for LVLMs directly optimized for on-device latency. In this paper, we present UltraViT, a vision encoder for LVLMs, explicitly designed and optimized for"},"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":"2607.23373","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-07-25T21:31:40Z","cross_cats_sorted":[],"title_canon_sha256":"7e18790648d2ca2dec1e2b86f6b684a397894bea5107ee3a8abdc1856bb235e8","abstract_canon_sha256":"095bb66644e4a99af2e9207f99595117e5a5e7ae4f4298f08daf55dd9cf71e45"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T01:22:49.644645Z","signature_b64":"Kz5EhrdYCmkzeu3NIPEpFGQJCTM/gwpdFgEuxC2qSm7XiVruzsLSQ+gD9qHtzjYorYacpBkxtd9XNIMyvwahAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1483e1d3c5bb561be465dd71ff588ccd9c6ca4732998ecd8fb7229898de6027f","last_reissued_at":"2026-07-28T01:22:49.643831Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T01:22:49.643831Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"UltraViT: Latency-Optimized On-device Vision Encoder for Large Vision-Language Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Adrian Bulat, Alberto Baldrati, Anestis Zaganidis, Georgios Tzimiropoulos, Hyeonuk Kim, Ioannis Maniadis Metaxas, Yassine Ouali","submitted_at":"2026-07-25T21:31:40Z","abstract_excerpt":"Large Vision-Language Models (LVLMs) remain bottlenecked by massive computational footprints, precluding their deployment on resource-constrained edge devices. While efforts to compress LVLMs focus heavily on vision token reduction or smaller language models, the vision encoder is largely overlooked, typically deployed as a monolithic, computationally heavy feature extractor. Moreover, there is no previous effort that designs a vision encoder for LVLMs directly optimized for on-device latency. In this paper, we present UltraViT, a vision encoder for LVLMs, explicitly designed and optimized for"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.23373","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/2607.23373/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":"2607.23373","created_at":"2026-07-28T01:22:49.644239+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.23373v1","created_at":"2026-07-28T01:22:49.644239+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.23373","created_at":"2026-07-28T01:22:49.644239+00:00"},{"alias_kind":"pith_short_12","alias_value":"CSB6DU6FXNLB","created_at":"2026-07-28T01:22:49.644239+00:00"},{"alias_kind":"pith_short_16","alias_value":"CSB6DU6FXNLBXZDF","created_at":"2026-07-28T01:22:49.644239+00:00"},{"alias_kind":"pith_short_8","alias_value":"CSB6DU6F","created_at":"2026-07-28T01:22:49.644239+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/CSB6DU6FXNLBXZDF3VY76WEMZW","json":"https://pith.science/pith/CSB6DU6FXNLBXZDF3VY76WEMZW.json","graph_json":"https://pith.science/api/pith-number/CSB6DU6FXNLBXZDF3VY76WEMZW/graph.json","events_json":"https://pith.science/api/pith-number/CSB6DU6FXNLBXZDF3VY76WEMZW/events.json","paper":"https://pith.science/paper/CSB6DU6F"},"agent_actions":{"view_html":"https://pith.science/pith/CSB6DU6FXNLBXZDF3VY76WEMZW","download_json":"https://pith.science/pith/CSB6DU6FXNLBXZDF3VY76WEMZW.json","view_paper":"https://pith.science/paper/CSB6DU6F","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.23373&json=true","fetch_graph":"https://pith.science/api/pith-number/CSB6DU6FXNLBXZDF3VY76WEMZW/graph.json","fetch_events":"https://pith.science/api/pith-number/CSB6DU6FXNLBXZDF3VY76WEMZW/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CSB6DU6FXNLBXZDF3VY76WEMZW/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CSB6DU6FXNLBXZDF3VY76WEMZW/action/storage_attestation","attest_author":"https://pith.science/pith/CSB6DU6FXNLBXZDF3VY76WEMZW/action/author_attestation","sign_citation":"https://pith.science/pith/CSB6DU6FXNLBXZDF3VY76WEMZW/action/citation_signature","submit_replication":"https://pith.science/pith/CSB6DU6FXNLBXZDF3VY76WEMZW/action/replication_record"}},"created_at":"2026-07-28T01:22:49.644239+00:00","updated_at":"2026-07-28T01:22:49.644239+00:00"}