{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:4EN75TSUPPQOTQXKYRHQFQTQZL","short_pith_number":"pith:4EN75TSU","schema_version":"1.0","canonical_sha256":"e11bfece547be0e9c2eac44f02c270cac13d5d651f2191fd160d937ce6f5bd02","source":{"kind":"arxiv","id":"2509.01819","version":1},"attestation_state":"computed","paper":{"title":"ManiFlow: A General Robot Manipulation Policy via Consistency Flow Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Ankit Goyal, Dieter Fox, Ge Yan, Jiyue Zhu, Marius Memmel, Ranjay Krishna, Ri-Zhao Qiu, Shiqi Yang, Xiaolong Wang, Xuxin Cheng, Yuquan Deng","submitted_at":"2025-09-01T22:50:55Z","abstract_excerpt":"This paper introduces ManiFlow, a visuomotor imitation learning policy for general robot manipulation that generates precise, high-dimensional actions conditioned on diverse visual, language and proprioceptive inputs. We leverage flow matching with consistency training to enable high-quality dexterous action generation in just 1-2 inference steps. To handle diverse input modalities efficiently, we propose DiT-X, a diffusion transformer architecture with adaptive cross-attention and AdaLN-Zero conditioning that enables fine-grained feature interactions between action tokens and multi-modal obse"},"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":"2509.01819","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2025-09-01T22:50:55Z","cross_cats_sorted":[],"title_canon_sha256":"8eb1408dbc9732750b379d2d12beb8166f03dcfb0bdd17f13ee046da4cb9120c","abstract_canon_sha256":"bcd894b5a23d7e555bab9deddba16a3afb6378f9472b30e09668ef110e54c9d3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:03:09.387177Z","signature_b64":"g5lUNMNAq/NDoRkE67sdiTbcyYF9jv0+21yoCJeQi2LnTUavI79iGaF/Nzu82x0T0r45z8/thdxCGOqbKveKBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e11bfece547be0e9c2eac44f02c270cac13d5d651f2191fd160d937ce6f5bd02","last_reissued_at":"2026-07-05T12:03:09.386706Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:03:09.386706Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"ManiFlow: A General Robot Manipulation Policy via Consistency Flow Training","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Ankit Goyal, Dieter Fox, Ge Yan, Jiyue Zhu, Marius Memmel, Ranjay Krishna, Ri-Zhao Qiu, Shiqi Yang, Xiaolong Wang, Xuxin Cheng, Yuquan Deng","submitted_at":"2025-09-01T22:50:55Z","abstract_excerpt":"This paper introduces ManiFlow, a visuomotor imitation learning policy for general robot manipulation that generates precise, high-dimensional actions conditioned on diverse visual, language and proprioceptive inputs. We leverage flow matching with consistency training to enable high-quality dexterous action generation in just 1-2 inference steps. To handle diverse input modalities efficiently, we propose DiT-X, a diffusion transformer architecture with adaptive cross-attention and AdaLN-Zero conditioning that enables fine-grained feature interactions between action tokens and multi-modal obse"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.01819","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/2509.01819/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":"2509.01819","created_at":"2026-07-05T12:03:09.386774+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.01819v1","created_at":"2026-07-05T12:03:09.386774+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.01819","created_at":"2026-07-05T12:03:09.386774+00:00"},{"alias_kind":"pith_short_12","alias_value":"4EN75TSUPPQO","created_at":"2026-07-05T12:03:09.386774+00:00"},{"alias_kind":"pith_short_16","alias_value":"4EN75TSUPPQOTQXK","created_at":"2026-07-05T12:03:09.386774+00:00"},{"alias_kind":"pith_short_8","alias_value":"4EN75TSU","created_at":"2026-07-05T12:03:09.386774+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":10,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.22480","citing_title":"ARP: Enhancing Quantized Skill Abstractions via Visual Alignment and Iterative Refinement for Robotic Manipulation","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2606.12499","citing_title":"Action-Effect Memory Pretraining for Robot Manipulation","ref_index":19,"is_internal_anchor":false},{"citing_arxiv_id":"2606.10517","citing_title":"LAFP: Preserving Latent Action Structure in Latent Policy Learning via Flow Matching","ref_index":27,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06461","citing_title":"Flow-based Policy Adaptation without Policy Updates","ref_index":36,"is_internal_anchor":false},{"citing_arxiv_id":"2606.03834","citing_title":"Let the Dynamics Flow: Stable Flow Matching Dynamical Systems","ref_index":41,"is_internal_anchor":false},{"citing_arxiv_id":"2606.01072","citing_title":"Expanding Spatial and Temporal Context for Robotic Imitation Learning With Scene Graphs","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2606.27475","citing_title":"Support-Constrained RL Enables Real-World Policy Improvement without Real-World Experience","ref_index":6,"is_internal_anchor":false},{"citing_arxiv_id":"2605.14598","citing_title":"DSSP: Diffusion State Space Policy with Full-History Encoding","ref_index":57,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15281","citing_title":"R3D: Revisiting 3D Policy Learning","ref_index":44,"is_internal_anchor":false},{"citing_arxiv_id":"2604.15569","citing_title":"ShapeGen: Robotic Data Generation for Category-Level Manipulation","ref_index":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4EN75TSUPPQOTQXKYRHQFQTQZL","json":"https://pith.science/pith/4EN75TSUPPQOTQXKYRHQFQTQZL.json","graph_json":"https://pith.science/api/pith-number/4EN75TSUPPQOTQXKYRHQFQTQZL/graph.json","events_json":"https://pith.science/api/pith-number/4EN75TSUPPQOTQXKYRHQFQTQZL/events.json","paper":"https://pith.science/paper/4EN75TSU"},"agent_actions":{"view_html":"https://pith.science/pith/4EN75TSUPPQOTQXKYRHQFQTQZL","download_json":"https://pith.science/pith/4EN75TSUPPQOTQXKYRHQFQTQZL.json","view_paper":"https://pith.science/paper/4EN75TSU","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.01819&json=true","fetch_graph":"https://pith.science/api/pith-number/4EN75TSUPPQOTQXKYRHQFQTQZL/graph.json","fetch_events":"https://pith.science/api/pith-number/4EN75TSUPPQOTQXKYRHQFQTQZL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4EN75TSUPPQOTQXKYRHQFQTQZL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4EN75TSUPPQOTQXKYRHQFQTQZL/action/storage_attestation","attest_author":"https://pith.science/pith/4EN75TSUPPQOTQXKYRHQFQTQZL/action/author_attestation","sign_citation":"https://pith.science/pith/4EN75TSUPPQOTQXKYRHQFQTQZL/action/citation_signature","submit_replication":"https://pith.science/pith/4EN75TSUPPQOTQXKYRHQFQTQZL/action/replication_record"}},"created_at":"2026-07-05T12:03:09.386774+00:00","updated_at":"2026-07-05T12:03:09.386774+00:00"}