{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:47LWLVJSY5TLFGWMTJKTM2KSRZ","short_pith_number":"pith:47LWLVJS","schema_version":"1.0","canonical_sha256":"e7d765d532c766b29acc9a553669528e63723e1518ecd819a169befdef1e49ce","source":{"kind":"arxiv","id":"2508.06804","version":1},"attestation_state":"computed","paper":{"title":"D3P: Dynamic Denoising Diffusion Policy via Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Chao Yu, Feng Gao, Shu-Ang Yu, Yi Wu, Yu Wang","submitted_at":"2025-08-09T03:27:21Z","abstract_excerpt":"Diffusion policies excel at learning complex action distributions for robotic visuomotor tasks, yet their iterative denoising process poses a major bottleneck for real-time deployment. Existing acceleration methods apply a fixed number of denoising steps per action, implicitly treating all actions as equally important. However, our experiments reveal that robotic tasks often contain a mix of \\emph{crucial} and \\emph{routine} actions, which differ in their impact on task success. Motivated by this finding, we propose \\textbf{D}ynamic \\textbf{D}enoising \\textbf{D}iffusion \\textbf{P}olicy \\textbf"},"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":"2508.06804","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-08-09T03:27:21Z","cross_cats_sorted":[],"title_canon_sha256":"ac71068821cc0b29c914026e46f7c80ff124c967898903f0cb16c1ed438485e3","abstract_canon_sha256":"237dfe30d69b568344acf13399f9be1cdb740a69363699b549ce57a49e63da6e"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:51:35.684219Z","signature_b64":"sdJuaKnSEID0akuvT8EY0pCcMgQNOM6VNXIvH272RqwUQLfCu0OWy9g6jIPlMASU12FxNgY8OxnepDgH6iWCAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e7d765d532c766b29acc9a553669528e63723e1518ecd819a169befdef1e49ce","last_reissued_at":"2026-07-05T11:51:35.683522Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:51:35.683522Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"D3P: Dynamic Denoising Diffusion Policy via Reinforcement Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Chao Yu, Feng Gao, Shu-Ang Yu, Yi Wu, Yu Wang","submitted_at":"2025-08-09T03:27:21Z","abstract_excerpt":"Diffusion policies excel at learning complex action distributions for robotic visuomotor tasks, yet their iterative denoising process poses a major bottleneck for real-time deployment. Existing acceleration methods apply a fixed number of denoising steps per action, implicitly treating all actions as equally important. However, our experiments reveal that robotic tasks often contain a mix of \\emph{crucial} and \\emph{routine} actions, which differ in their impact on task success. Motivated by this finding, we propose \\textbf{D}ynamic \\textbf{D}enoising \\textbf{D}iffusion \\textbf{P}olicy \\textbf"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2508.06804","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/2508.06804/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":"2508.06804","created_at":"2026-07-05T11:51:35.683594+00:00"},{"alias_kind":"arxiv_version","alias_value":"2508.06804v1","created_at":"2026-07-05T11:51:35.683594+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2508.06804","created_at":"2026-07-05T11:51:35.683594+00:00"},{"alias_kind":"pith_short_12","alias_value":"47LWLVJSY5TL","created_at":"2026-07-05T11:51:35.683594+00:00"},{"alias_kind":"pith_short_16","alias_value":"47LWLVJSY5TLFGWM","created_at":"2026-07-05T11:51:35.683594+00:00"},{"alias_kind":"pith_short_8","alias_value":"47LWLVJS","created_at":"2026-07-05T11:51:35.683594+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06049","citing_title":"L-SDPPO: Policy Optimization of Spiking Diffusion Policy for Intra-vehicular Robotic Manipulation","ref_index":35,"is_internal_anchor":false},{"citing_arxiv_id":"2605.31460","citing_title":"On-Device Robotic Planning: Eliminating Inference Redundancy for Efficient Decision-Making","ref_index":64,"is_internal_anchor":false},{"citing_arxiv_id":"2606.31132","citing_title":"ELASTIC: Efficiently Learning to Adaptively Scale Test-Time Compute for Generative Control Policies","ref_index":15,"is_internal_anchor":false},{"citing_arxiv_id":"2605.27947","citing_title":"SANTS: A State-Adaptive Scheduler for World Action Models","ref_index":16,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08733","citing_title":"Generative Actor-Critic with Soft Bridge Policies","ref_index":16,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/47LWLVJSY5TLFGWMTJKTM2KSRZ","json":"https://pith.science/pith/47LWLVJSY5TLFGWMTJKTM2KSRZ.json","graph_json":"https://pith.science/api/pith-number/47LWLVJSY5TLFGWMTJKTM2KSRZ/graph.json","events_json":"https://pith.science/api/pith-number/47LWLVJSY5TLFGWMTJKTM2KSRZ/events.json","paper":"https://pith.science/paper/47LWLVJS"},"agent_actions":{"view_html":"https://pith.science/pith/47LWLVJSY5TLFGWMTJKTM2KSRZ","download_json":"https://pith.science/pith/47LWLVJSY5TLFGWMTJKTM2KSRZ.json","view_paper":"https://pith.science/paper/47LWLVJS","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2508.06804&json=true","fetch_graph":"https://pith.science/api/pith-number/47LWLVJSY5TLFGWMTJKTM2KSRZ/graph.json","fetch_events":"https://pith.science/api/pith-number/47LWLVJSY5TLFGWMTJKTM2KSRZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/47LWLVJSY5TLFGWMTJKTM2KSRZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/47LWLVJSY5TLFGWMTJKTM2KSRZ/action/storage_attestation","attest_author":"https://pith.science/pith/47LWLVJSY5TLFGWMTJKTM2KSRZ/action/author_attestation","sign_citation":"https://pith.science/pith/47LWLVJSY5TLFGWMTJKTM2KSRZ/action/citation_signature","submit_replication":"https://pith.science/pith/47LWLVJSY5TLFGWMTJKTM2KSRZ/action/replication_record"}},"created_at":"2026-07-05T11:51:35.683594+00:00","updated_at":"2026-07-05T11:51:35.683594+00:00"}