{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZACJVEXA55LVPZKMIHK2HAEVRO","short_pith_number":"pith:ZACJVEXA","schema_version":"1.0","canonical_sha256":"c8049a92e0ef5757e54c41d5a380958b93524231c30e57abfae3f5637297d9dc","source":{"kind":"arxiv","id":"2607.20293","version":1},"attestation_state":"computed","paper":{"title":"Evolving Cache Schedules for Fast Diffusion Policy Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Di Wang, Fei Cheng, Kangye Ji, Siying Wang","submitted_at":"2026-07-22T15:40:11Z","abstract_excerpt":"Diffusion policies achieve strong visuomotor control by iteratively denoising action chunks, but repeated denoising makes real-time deployment computationally demanding. Cache-based methods reduce inference cost by reusing intermediate activations, but existing training-free schedules typically allocate computation uniformly across blocks, ignoring heterogeneous redundancy across blocks and leading to a suboptimal performance-efficiency trade-off. To bridge this gap, we introduce Evolving Cache Schedules (EVO), a training-free acceleration framework that globally schedules cache refreshes via "},"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":"2607.20293","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-07-22T15:40:11Z","cross_cats_sorted":[],"title_canon_sha256":"eb45d6da280a07ca9f4689f1cf67256110fd5be248359ee2c5a1fe9c76092860","abstract_canon_sha256":"827c607611fb7a4330c73052f3dd08b57689c26f43301d4eca21dc4bc8a0365c"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-23T01:25:12.797997Z","signature_b64":"Q7WA92qWrvlIZl6QndR4Fh+TOZOg3Czz/7zYifp7Spvo/pYYE1QO0nBq0ceiIJnWIJaEMmByKCulNOP02Hp7Ag==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c8049a92e0ef5757e54c41d5a380958b93524231c30e57abfae3f5637297d9dc","last_reissued_at":"2026-07-23T01:25:12.797130Z","signature_status":"signed_v1","first_computed_at":"2026-07-23T01:25:12.797130Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evolving Cache Schedules for Fast Diffusion Policy Inference","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Di Wang, Fei Cheng, Kangye Ji, Siying Wang","submitted_at":"2026-07-22T15:40:11Z","abstract_excerpt":"Diffusion policies achieve strong visuomotor control by iteratively denoising action chunks, but repeated denoising makes real-time deployment computationally demanding. Cache-based methods reduce inference cost by reusing intermediate activations, but existing training-free schedules typically allocate computation uniformly across blocks, ignoring heterogeneous redundancy across blocks and leading to a suboptimal performance-efficiency trade-off. To bridge this gap, we introduce Evolving Cache Schedules (EVO), a training-free acceleration framework that globally schedules cache refreshes via "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.20293","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/2607.20293/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":"2607.20293","created_at":"2026-07-23T01:25:12.797580+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.20293v1","created_at":"2026-07-23T01:25:12.797580+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.20293","created_at":"2026-07-23T01:25:12.797580+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZACJVEXA55LV","created_at":"2026-07-23T01:25:12.797580+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZACJVEXA55LVPZKM","created_at":"2026-07-23T01:25:12.797580+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZACJVEXA","created_at":"2026-07-23T01:25:12.797580+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/ZACJVEXA55LVPZKMIHK2HAEVRO","json":"https://pith.science/pith/ZACJVEXA55LVPZKMIHK2HAEVRO.json","graph_json":"https://pith.science/api/pith-number/ZACJVEXA55LVPZKMIHK2HAEVRO/graph.json","events_json":"https://pith.science/api/pith-number/ZACJVEXA55LVPZKMIHK2HAEVRO/events.json","paper":"https://pith.science/paper/ZACJVEXA"},"agent_actions":{"view_html":"https://pith.science/pith/ZACJVEXA55LVPZKMIHK2HAEVRO","download_json":"https://pith.science/pith/ZACJVEXA55LVPZKMIHK2HAEVRO.json","view_paper":"https://pith.science/paper/ZACJVEXA","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.20293&json=true","fetch_graph":"https://pith.science/api/pith-number/ZACJVEXA55LVPZKMIHK2HAEVRO/graph.json","fetch_events":"https://pith.science/api/pith-number/ZACJVEXA55LVPZKMIHK2HAEVRO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZACJVEXA55LVPZKMIHK2HAEVRO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZACJVEXA55LVPZKMIHK2HAEVRO/action/storage_attestation","attest_author":"https://pith.science/pith/ZACJVEXA55LVPZKMIHK2HAEVRO/action/author_attestation","sign_citation":"https://pith.science/pith/ZACJVEXA55LVPZKMIHK2HAEVRO/action/citation_signature","submit_replication":"https://pith.science/pith/ZACJVEXA55LVPZKMIHK2HAEVRO/action/replication_record"}},"created_at":"2026-07-23T01:25:12.797580+00:00","updated_at":"2026-07-23T01:25:12.797580+00:00"}