{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:ZA5PKBLJTYOXHLD4I775HIOXDM","short_pith_number":"pith:ZA5PKBLJ","schema_version":"1.0","canonical_sha256":"c83af505699e1d73ac7c47ffd3a1d71b272315f008ed7349d7c9ab60340e8ecb","source":{"kind":"arxiv","id":"2408.17355","version":4},"attestation_state":"computed","paper":{"title":"Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Annie Xie, Chelsea Finn, Jubayer Ibn Hamid, Maximilian Du, Yoonho Lee, Yuejiang Liu","submitted_at":"2024-08-30T15:39:34Z","abstract_excerpt":"Predicting and executing a sequence of actions without intermediate replanning, known as action chunking, is increasingly used in robot learning from human demonstrations. Yet, its effects on the learned policy remain inconsistent: some studies find it crucial for achieving strong results, while others observe decreased performance. In this paper, we first dissect how action chunking impacts the divergence between a learner and a demonstrator. We find that action chunking allows the learner to better capture the temporal dependencies in demonstrations but at the cost of reduced reactivity to u"},"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":"2408.17355","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2024-08-30T15:39:34Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"6899a3d82e545dd390d1b294367ad2e5d53cbd3c6ed366d74c4f11f52a5ffeca","abstract_canon_sha256":"771cdac095b15566903dd8a79c6c06a469d3fbe391555ae1bf9b3083debcf572"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:53:47.213033Z","signature_b64":"Of5BJNnirZEW8AhuKZIpaat8gfvx81F6dnesfyf+AbZKMEfqi+lWXlANhmdHcbUDQ4jDxm+KHhoeuF0MOkQRCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c83af505699e1d73ac7c47ffd3a1d71b272315f008ed7349d7c9ab60340e8ecb","last_reissued_at":"2026-07-05T10:53:47.212478Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:53:47.212478Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Annie Xie, Chelsea Finn, Jubayer Ibn Hamid, Maximilian Du, Yoonho Lee, Yuejiang Liu","submitted_at":"2024-08-30T15:39:34Z","abstract_excerpt":"Predicting and executing a sequence of actions without intermediate replanning, known as action chunking, is increasingly used in robot learning from human demonstrations. Yet, its effects on the learned policy remain inconsistent: some studies find it crucial for achieving strong results, while others observe decreased performance. In this paper, we first dissect how action chunking impacts the divergence between a learner and a demonstrator. We find that action chunking allows the learner to better capture the temporal dependencies in demonstrations but at the cost of reduced reactivity to u"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2408.17355","kind":"arxiv","version":4},"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/2408.17355/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":"2408.17355","created_at":"2026-07-05T10:53:47.212538+00:00"},{"alias_kind":"arxiv_version","alias_value":"2408.17355v4","created_at":"2026-07-05T10:53:47.212538+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2408.17355","created_at":"2026-07-05T10:53:47.212538+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZA5PKBLJTYOX","created_at":"2026-07-05T10:53:47.212538+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZA5PKBLJTYOXHLD4","created_at":"2026-07-05T10:53:47.212538+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZA5PKBLJ","created_at":"2026-07-05T10:53:47.212538+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":13,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2607.05391","citing_title":"LLM-as-a-Verifier: A General-Purpose Verification Framework","ref_index":87,"is_internal_anchor":true},{"citing_arxiv_id":"2606.26588","citing_title":"Inference-Time Robot Behavior Steering through Physically-Aware Reconfiguration of Task-Structure","ref_index":58,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19774","citing_title":"Start Right, Arrive Right: Asynchronous Execution via Initial Noise Selection","ref_index":9,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01804","citing_title":"VLA-Corrector: Lightweight Detect-and-Correct Inference for Adaptive Action Horizon","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08094","citing_title":"vla.cpp: A Unified Inference Runtime for Vision-Language-Action Models","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2607.01051","citing_title":"AutoSpeed: Annotation-Free Stage-Adaptive Motion Speed Learning for Robot Manipulation","ref_index":24,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25050","citing_title":"DiscreteRTC: Discrete Diffusion Policies are Natural Asynchronous Executors","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2605.24433","citing_title":"Smoother Action Chunking Flow Policy via Prior-Corrected Orthogonal Trust-Region Guidance","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"2602.12978","citing_title":"Learning Native Continuation for Action Chunking Flow Policies","ref_index":26,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17486","citing_title":"DyGRO-VLA: Cross-Task Scaling of Vision-Language-Action Models via Dynamic Grouped Residual Optimization","ref_index":82,"is_internal_anchor":false},{"citing_arxiv_id":"2506.07339","citing_title":"Real-Time Execution of Action Chunking Flow Policies","ref_index":39,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25050","citing_title":"DiscreteRTC: Discrete Diffusion Policies are Natural Asynchronous Executors","ref_index":34,"is_internal_anchor":false},{"citing_arxiv_id":"2502.19645","citing_title":"Fine-Tuning Vision-Language-Action Models: Optimizing Speed and Success","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZA5PKBLJTYOXHLD4I775HIOXDM","json":"https://pith.science/pith/ZA5PKBLJTYOXHLD4I775HIOXDM.json","graph_json":"https://pith.science/api/pith-number/ZA5PKBLJTYOXHLD4I775HIOXDM/graph.json","events_json":"https://pith.science/api/pith-number/ZA5PKBLJTYOXHLD4I775HIOXDM/events.json","paper":"https://pith.science/paper/ZA5PKBLJ"},"agent_actions":{"view_html":"https://pith.science/pith/ZA5PKBLJTYOXHLD4I775HIOXDM","download_json":"https://pith.science/pith/ZA5PKBLJTYOXHLD4I775HIOXDM.json","view_paper":"https://pith.science/paper/ZA5PKBLJ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2408.17355&json=true","fetch_graph":"https://pith.science/api/pith-number/ZA5PKBLJTYOXHLD4I775HIOXDM/graph.json","fetch_events":"https://pith.science/api/pith-number/ZA5PKBLJTYOXHLD4I775HIOXDM/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZA5PKBLJTYOXHLD4I775HIOXDM/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZA5PKBLJTYOXHLD4I775HIOXDM/action/storage_attestation","attest_author":"https://pith.science/pith/ZA5PKBLJTYOXHLD4I775HIOXDM/action/author_attestation","sign_citation":"https://pith.science/pith/ZA5PKBLJTYOXHLD4I775HIOXDM/action/citation_signature","submit_replication":"https://pith.science/pith/ZA5PKBLJTYOXHLD4I775HIOXDM/action/replication_record"}},"created_at":"2026-07-05T10:53:47.212538+00:00","updated_at":"2026-07-05T10:53:47.212538+00:00"}