{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:JXNS6OPVDLNN2URBE5HVAYACBG","short_pith_number":"pith:JXNS6OPV","schema_version":"1.0","canonical_sha256":"4ddb2f39f51adadd5221274f50600209858cc4b2cb54e322290d32dc02584afd","source":{"kind":"arxiv","id":"2601.02295","version":2},"attestation_state":"computed","paper":{"title":"CycleVLA: Proactive Self-Correcting Vision-Language-Action Models via Subtask Backtracking and Minimum Bayes Risk Decoding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Andrew Markham, Bill Byrne, Chenyang Ma, Guangyu Yang, Ioannis Havoutis, Jiuming Liu, Kai Lu, Niki Trigoni, Shitong Xu","submitted_at":"2026-01-05T17:31:01Z","abstract_excerpt":"Current work on robot failure detection and correction typically operates in a post hoc manner, analyzing errors and applying corrections only after failures occur. This work introduces CycleVLA, a system that equips Vision-Language-Action models (VLAs) with proactive self-correction, the capability to anticipate incipient failures and recover before they fully manifest during execution. CycleVLA achieves this by integrating a progress-aware VLA that flags critical subtask transition points where failures most frequently occur, a VLM-based failure predictor and planner that triggers subtask ba"},"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":"2601.02295","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-01-05T17:31:01Z","cross_cats_sorted":[],"title_canon_sha256":"4aab11f115af720bf52bc68c83b68803f91969295edfd4a0596a920212a1648f","abstract_canon_sha256":"e1bb0975ab41530e08350b245143e7f694351360cf0d87ca5f7899cc4d2fc705"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-29T01:25:32.846300Z","signature_b64":"oX/6d1wDRWNgoRJvkJwhx/ZYP7MLVK8vgFAl9c6mzwGTN+I6l9REHGGJyqCKM9gZE69zZxIvl5aXRdNrQHsyCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"4ddb2f39f51adadd5221274f50600209858cc4b2cb54e322290d32dc02584afd","last_reissued_at":"2026-07-29T01:25:32.845328Z","signature_status":"signed_v1","first_computed_at":"2026-07-29T01:25:32.845328Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"CycleVLA: Proactive Self-Correcting Vision-Language-Action Models via Subtask Backtracking and Minimum Bayes Risk Decoding","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Andrew Markham, Bill Byrne, Chenyang Ma, Guangyu Yang, Ioannis Havoutis, Jiuming Liu, Kai Lu, Niki Trigoni, Shitong Xu","submitted_at":"2026-01-05T17:31:01Z","abstract_excerpt":"Current work on robot failure detection and correction typically operates in a post hoc manner, analyzing errors and applying corrections only after failures occur. This work introduces CycleVLA, a system that equips Vision-Language-Action models (VLAs) with proactive self-correction, the capability to anticipate incipient failures and recover before they fully manifest during execution. CycleVLA achieves this by integrating a progress-aware VLA that flags critical subtask transition points where failures most frequently occur, a VLM-based failure predictor and planner that triggers subtask ba"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2601.02295","kind":"arxiv","version":2},"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/2601.02295/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":"2601.02295","created_at":"2026-07-29T01:25:32.845791+00:00"},{"alias_kind":"arxiv_version","alias_value":"2601.02295v2","created_at":"2026-07-29T01:25:32.845791+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2601.02295","created_at":"2026-07-29T01:25:32.845791+00:00"},{"alias_kind":"pith_short_12","alias_value":"JXNS6OPVDLNN","created_at":"2026-07-29T01:25:32.845791+00:00"},{"alias_kind":"pith_short_16","alias_value":"JXNS6OPVDLNN2URB","created_at":"2026-07-29T01:25:32.845791+00:00"},{"alias_kind":"pith_short_8","alias_value":"JXNS6OPV","created_at":"2026-07-29T01:25:32.845791+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":4,"sample":[{"citing_arxiv_id":"2607.06534","citing_title":"CAIRN: Cross-Room 3D Scene Understanding with Topology-Aware Large Multimodal Models","ref_index":36,"is_internal_anchor":true},{"citing_arxiv_id":"2606.21386","citing_title":"VLA-FAIL: Efficient Task Failure Detection for Finetuned Vision-Language-Action Models","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2606.09740","citing_title":"ProbeAct: Probe-Guided Training-Free Failure Recovery in Vision-Language-Action Models","ref_index":14,"is_internal_anchor":true},{"citing_arxiv_id":"2607.01212","citing_title":"FurnitureVLA: Learning Long-Horizon Bimanual Furniture Assembly with Vision-Language-Action Model","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/JXNS6OPVDLNN2URBE5HVAYACBG","json":"https://pith.science/pith/JXNS6OPVDLNN2URBE5HVAYACBG.json","graph_json":"https://pith.science/api/pith-number/JXNS6OPVDLNN2URBE5HVAYACBG/graph.json","events_json":"https://pith.science/api/pith-number/JXNS6OPVDLNN2URBE5HVAYACBG/events.json","paper":"https://pith.science/paper/JXNS6OPV"},"agent_actions":{"view_html":"https://pith.science/pith/JXNS6OPVDLNN2URBE5HVAYACBG","download_json":"https://pith.science/pith/JXNS6OPVDLNN2URBE5HVAYACBG.json","view_paper":"https://pith.science/paper/JXNS6OPV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2601.02295&json=true","fetch_graph":"https://pith.science/api/pith-number/JXNS6OPVDLNN2URBE5HVAYACBG/graph.json","fetch_events":"https://pith.science/api/pith-number/JXNS6OPVDLNN2URBE5HVAYACBG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/JXNS6OPVDLNN2URBE5HVAYACBG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/JXNS6OPVDLNN2URBE5HVAYACBG/action/storage_attestation","attest_author":"https://pith.science/pith/JXNS6OPVDLNN2URBE5HVAYACBG/action/author_attestation","sign_citation":"https://pith.science/pith/JXNS6OPVDLNN2URBE5HVAYACBG/action/citation_signature","submit_replication":"https://pith.science/pith/JXNS6OPVDLNN2URBE5HVAYACBG/action/replication_record"}},"created_at":"2026-07-29T01:25:32.845791+00:00","updated_at":"2026-07-29T01:25:32.845791+00:00"}