{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:I4RJDGRH2KSBNPIPJZ2VSOVH5R","short_pith_number":"pith:I4RJDGRH","schema_version":"1.0","canonical_sha256":"4722919a27d2a416bd0f4e75593aa7ec4a9ff1fabca7d280063221125081f35d","source":{"kind":"arxiv","id":"2607.24424","version":1},"attestation_state":"computed","paper":{"title":"MAViE: A Multi-scale Adaptive Vision Encoder for Fine-grained Visual Perception and Efficient Multimodal Reasoning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Shaofei Lei","submitted_at":"2026-07-27T13:36:08Z","abstract_excerpt":"Vision-language models commonly project all tokens produced by a pretrained vision encoder into a large language model. However, final-layer features can discard text, local attributes, and spatial relationships, while high-resolution inputs substantially increase context length and inference latency. We introduce \\method, a Multi-scale Adaptive Vision Encoder. \\method uses position-dependent gates to fuse shallow, intermediate, and deep features from a vision Transformer, preserving global semantics while enhancing edges, text, and local structure. It then performs question-conditioned token "},"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":"2607.24424","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-27T13:36:08Z","cross_cats_sorted":[],"title_canon_sha256":"3c124284d243da88a8ad0f070e5f1f6b8107251c528b23d1115d48ed79d1e819","abstract_canon_sha256":"4c12c80f81230af33fb53591cde92f2e090fbdfb645ee313c8e9af48706fcb7e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-28T02:24:02.779348Z","signature_b64":"PNePCVkYPMhxMj1jV4m1OlqA/aG5WSS9+ZOyt4Zt4r7R8Xz5iDzW1QTAO310GtyUYB7b8FkQZJz0oUZ13vxQDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4722919a27d2a416bd0f4e75593aa7ec4a9ff1fabca7d280063221125081f35d","last_reissued_at":"2026-07-28T02:24:02.778473Z","signature_status":"signed_v1","first_computed_at":"2026-07-28T02:24:02.778473Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"MAViE: A Multi-scale Adaptive Vision Encoder for Fine-grained Visual Perception and Efficient Multimodal Reasoning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Shaofei Lei","submitted_at":"2026-07-27T13:36:08Z","abstract_excerpt":"Vision-language models commonly project all tokens produced by a pretrained vision encoder into a large language model. However, final-layer features can discard text, local attributes, and spatial relationships, while high-resolution inputs substantially increase context length and inference latency. We introduce \\method, a Multi-scale Adaptive Vision Encoder. \\method uses position-dependent gates to fuse shallow, intermediate, and deep features from a vision Transformer, preserving global semantics while enhancing edges, text, and local structure. It then performs question-conditioned token "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.24424","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/2607.24424/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":"2607.24424","created_at":"2026-07-28T02:24:02.778916+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.24424v1","created_at":"2026-07-28T02:24:02.778916+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.24424","created_at":"2026-07-28T02:24:02.778916+00:00"},{"alias_kind":"pith_short_12","alias_value":"I4RJDGRH2KSB","created_at":"2026-07-28T02:24:02.778916+00:00"},{"alias_kind":"pith_short_16","alias_value":"I4RJDGRH2KSBNPIP","created_at":"2026-07-28T02:24:02.778916+00:00"},{"alias_kind":"pith_short_8","alias_value":"I4RJDGRH","created_at":"2026-07-28T02:24:02.778916+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/I4RJDGRH2KSBNPIPJZ2VSOVH5R","json":"https://pith.science/pith/I4RJDGRH2KSBNPIPJZ2VSOVH5R.json","graph_json":"https://pith.science/api/pith-number/I4RJDGRH2KSBNPIPJZ2VSOVH5R/graph.json","events_json":"https://pith.science/api/pith-number/I4RJDGRH2KSBNPIPJZ2VSOVH5R/events.json","paper":"https://pith.science/paper/I4RJDGRH"},"agent_actions":{"view_html":"https://pith.science/pith/I4RJDGRH2KSBNPIPJZ2VSOVH5R","download_json":"https://pith.science/pith/I4RJDGRH2KSBNPIPJZ2VSOVH5R.json","view_paper":"https://pith.science/paper/I4RJDGRH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.24424&json=true","fetch_graph":"https://pith.science/api/pith-number/I4RJDGRH2KSBNPIPJZ2VSOVH5R/graph.json","fetch_events":"https://pith.science/api/pith-number/I4RJDGRH2KSBNPIPJZ2VSOVH5R/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/I4RJDGRH2KSBNPIPJZ2VSOVH5R/action/timestamp_anchor","attest_storage":"https://pith.science/pith/I4RJDGRH2KSBNPIPJZ2VSOVH5R/action/storage_attestation","attest_author":"https://pith.science/pith/I4RJDGRH2KSBNPIPJZ2VSOVH5R/action/author_attestation","sign_citation":"https://pith.science/pith/I4RJDGRH2KSBNPIPJZ2VSOVH5R/action/citation_signature","submit_replication":"https://pith.science/pith/I4RJDGRH2KSBNPIPJZ2VSOVH5R/action/replication_record"}},"created_at":"2026-07-28T02:24:02.778916+00:00","updated_at":"2026-07-28T02:24:02.778916+00:00"}