{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:IRKH44ADBVKAXEX5T3CL6TZZSV","short_pith_number":"pith:IRKH44AD","schema_version":"1.0","canonical_sha256":"44547e70030d540b92fd9ec4bf4f39956d639a29fb4dd0e43199feb902b39867","source":{"kind":"arxiv","id":"2605.23163","version":1},"attestation_state":"computed","paper":{"title":"Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Boris Ivanovic, Chengyue Wu, Daquan Zhou, Enze Xie, Jin Wang, Kewei Zhang, Langechuan Liu, Marco Pavone, Sensen Gao, Song Han, Songyang Han, Yulong Cao","submitted_at":"2026-05-22T02:31:32Z","abstract_excerpt":"End-to-end autonomous driving via Vision-Language-Action (VLA) models demands a precarious balance between high-fidelity trajectory planning and efficient inference. Existing paradigms typically fall short: autoregressive (AR) VLAs are memory-bandwidth-bound on edge hardware and prone to exposure-bias drift, while full-sequence diffusion models preclude KV-cache reuse and suffer from \"logical leakage\" that violates the fundamental perceive-then-plan causality. We present Fast-dDrive, a block-diffusion VLA that performs bidirectional refinement within semantic units while enforcing strict causa"},"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":"2605.23163","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-22T02:31:32Z","cross_cats_sorted":[],"title_canon_sha256":"b72b84961b855eba543cc71f4619ab944fc6fce03fb3b60ac2b455646a2130e1","abstract_canon_sha256":"7560a91d4a3f70144ce293672a41981b14ce82c971b9622cfa19483bd41099ce"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-25T02:01:41.117596Z","signature_b64":"loUvNyNj1z2nvy+4v7LCssmIm7Rq/6NegVW8M3Za7pq9zKjXILHUyoMCU56wr6Phq+G0E02fXZ8Kok/NV55mBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"44547e70030d540b92fd9ec4bf4f39956d639a29fb4dd0e43199feb902b39867","last_reissued_at":"2026-05-25T02:01:41.116936Z","signature_status":"signed_v1","first_computed_at":"2026-05-25T02:01:41.116936Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Boris Ivanovic, Chengyue Wu, Daquan Zhou, Enze Xie, Jin Wang, Kewei Zhang, Langechuan Liu, Marco Pavone, Sensen Gao, Song Han, Songyang Han, Yulong Cao","submitted_at":"2026-05-22T02:31:32Z","abstract_excerpt":"End-to-end autonomous driving via Vision-Language-Action (VLA) models demands a precarious balance between high-fidelity trajectory planning and efficient inference. Existing paradigms typically fall short: autoregressive (AR) VLAs are memory-bandwidth-bound on edge hardware and prone to exposure-bias drift, while full-sequence diffusion models preclude KV-cache reuse and suffer from \"logical leakage\" that violates the fundamental perceive-then-plan causality. We present Fast-dDrive, a block-diffusion VLA that performs bidirectional refinement within semantic units while enforcing strict causa"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.23163","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/2605.23163/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":"2605.23163","created_at":"2026-05-25T02:01:41.117038+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.23163v1","created_at":"2026-05-25T02:01:41.117038+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.23163","created_at":"2026-05-25T02:01:41.117038+00:00"},{"alias_kind":"pith_short_12","alias_value":"IRKH44ADBVKA","created_at":"2026-05-25T02:01:41.117038+00:00"},{"alias_kind":"pith_short_16","alias_value":"IRKH44ADBVKAXEX5","created_at":"2026-05-25T02:01:41.117038+00:00"},{"alias_kind":"pith_short_8","alias_value":"IRKH44AD","created_at":"2026-05-25T02:01:41.117038+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/IRKH44ADBVKAXEX5T3CL6TZZSV","json":"https://pith.science/pith/IRKH44ADBVKAXEX5T3CL6TZZSV.json","graph_json":"https://pith.science/api/pith-number/IRKH44ADBVKAXEX5T3CL6TZZSV/graph.json","events_json":"https://pith.science/api/pith-number/IRKH44ADBVKAXEX5T3CL6TZZSV/events.json","paper":"https://pith.science/paper/IRKH44AD"},"agent_actions":{"view_html":"https://pith.science/pith/IRKH44ADBVKAXEX5T3CL6TZZSV","download_json":"https://pith.science/pith/IRKH44ADBVKAXEX5T3CL6TZZSV.json","view_paper":"https://pith.science/paper/IRKH44AD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.23163&json=true","fetch_graph":"https://pith.science/api/pith-number/IRKH44ADBVKAXEX5T3CL6TZZSV/graph.json","fetch_events":"https://pith.science/api/pith-number/IRKH44ADBVKAXEX5T3CL6TZZSV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/IRKH44ADBVKAXEX5T3CL6TZZSV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/IRKH44ADBVKAXEX5T3CL6TZZSV/action/storage_attestation","attest_author":"https://pith.science/pith/IRKH44ADBVKAXEX5T3CL6TZZSV/action/author_attestation","sign_citation":"https://pith.science/pith/IRKH44ADBVKAXEX5T3CL6TZZSV/action/citation_signature","submit_replication":"https://pith.science/pith/IRKH44ADBVKAXEX5T3CL6TZZSV/action/replication_record"}},"created_at":"2026-05-25T02:01:41.117038+00:00","updated_at":"2026-05-25T02:01:41.117038+00:00"}