{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:XHBMBV67J3XYBMGU7B3QSVZKEL","short_pith_number":"pith:XHBMBV67","schema_version":"1.0","canonical_sha256":"b9c2c0d7df4eef80b0d4f87709572a22f348dd045c10fca541ce9a6684d4aeb8","source":{"kind":"arxiv","id":"2607.02612","version":1},"attestation_state":"computed","paper":{"title":"Fusion: A Framework for Unified Sequential Token AdaptatIon in VisiOn TraNsformers","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.AR","cs.DC","cs.LG"],"primary_cat":"cs.CV","authors_text":"Aravind Pradeep, Christian Herglotz, Mahdi Taheri, Samira Nazari","submitted_at":"2026-07-01T15:51:25Z","abstract_excerpt":"Vision Transformers achieve strong image classification accuracy but process all image regions with nearly the same computation, even when many regions are redundant or uninformative. Recent adaptive inference methods reduce this cost by selectively compressing tokens or terminating inference early, but combining these mechanisms often causes unstable intermediate representations and accuracy degradation. We introduce Fusion, a unified adaptive inference framework that coordinates token merging, early exiting, and token pruning through a simple staged design: tokens are merged first, confidenc"},"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":"2607.02612","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CV","submitted_at":"2026-07-01T15:51:25Z","cross_cats_sorted":["cs.AI","cs.AR","cs.DC","cs.LG"],"title_canon_sha256":"ab3e0ba65705b5afe70f39d1dc3c32aae1a1d07e3147da0f6853ca4557ca749d","abstract_canon_sha256":"d2d060ed6dd7292a8070e2f1cb79e0a4aea406962416efb2ba6542c026032bab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-07T00:16:03.424415Z","signature_b64":"d+RFYzai8+02e9dKelEsaryWu/o0st8zglhwcvhJNBRLwmzjktbPy3UWt5D3YkVB+R0h3JmSms0KOPUERdNsCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b9c2c0d7df4eef80b0d4f87709572a22f348dd045c10fca541ce9a6684d4aeb8","last_reissued_at":"2026-07-07T00:16:03.423884Z","signature_status":"signed_v1","first_computed_at":"2026-07-07T00:16:03.423884Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fusion: A Framework for Unified Sequential Token AdaptatIon in VisiOn TraNsformers","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI","cs.AR","cs.DC","cs.LG"],"primary_cat":"cs.CV","authors_text":"Aravind Pradeep, Christian Herglotz, Mahdi Taheri, Samira Nazari","submitted_at":"2026-07-01T15:51:25Z","abstract_excerpt":"Vision Transformers achieve strong image classification accuracy but process all image regions with nearly the same computation, even when many regions are redundant or uninformative. Recent adaptive inference methods reduce this cost by selectively compressing tokens or terminating inference early, but combining these mechanisms often causes unstable intermediate representations and accuracy degradation. We introduce Fusion, a unified adaptive inference framework that coordinates token merging, early exiting, and token pruning through a simple staged design: tokens are merged first, confidenc"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.02612","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/2607.02612/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":"2607.02612","created_at":"2026-07-07T00:16:03.423957+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.02612v1","created_at":"2026-07-07T00:16:03.423957+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.02612","created_at":"2026-07-07T00:16:03.423957+00:00"},{"alias_kind":"pith_short_12","alias_value":"XHBMBV67J3XY","created_at":"2026-07-07T00:16:03.423957+00:00"},{"alias_kind":"pith_short_16","alias_value":"XHBMBV67J3XYBMGU","created_at":"2026-07-07T00:16:03.423957+00:00"},{"alias_kind":"pith_short_8","alias_value":"XHBMBV67","created_at":"2026-07-07T00:16:03.423957+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/XHBMBV67J3XYBMGU7B3QSVZKEL","json":"https://pith.science/pith/XHBMBV67J3XYBMGU7B3QSVZKEL.json","graph_json":"https://pith.science/api/pith-number/XHBMBV67J3XYBMGU7B3QSVZKEL/graph.json","events_json":"https://pith.science/api/pith-number/XHBMBV67J3XYBMGU7B3QSVZKEL/events.json","paper":"https://pith.science/paper/XHBMBV67"},"agent_actions":{"view_html":"https://pith.science/pith/XHBMBV67J3XYBMGU7B3QSVZKEL","download_json":"https://pith.science/pith/XHBMBV67J3XYBMGU7B3QSVZKEL.json","view_paper":"https://pith.science/paper/XHBMBV67","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.02612&json=true","fetch_graph":"https://pith.science/api/pith-number/XHBMBV67J3XYBMGU7B3QSVZKEL/graph.json","fetch_events":"https://pith.science/api/pith-number/XHBMBV67J3XYBMGU7B3QSVZKEL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XHBMBV67J3XYBMGU7B3QSVZKEL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XHBMBV67J3XYBMGU7B3QSVZKEL/action/storage_attestation","attest_author":"https://pith.science/pith/XHBMBV67J3XYBMGU7B3QSVZKEL/action/author_attestation","sign_citation":"https://pith.science/pith/XHBMBV67J3XYBMGU7B3QSVZKEL/action/citation_signature","submit_replication":"https://pith.science/pith/XHBMBV67J3XYBMGU7B3QSVZKEL/action/replication_record"}},"created_at":"2026-07-07T00:16:03.423957+00:00","updated_at":"2026-07-07T00:16:03.423957+00:00"}