{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:DEUP5KJ272H4E34GO6K7ONZQ4M","short_pith_number":"pith:DEUP5KJ2","schema_version":"1.0","canonical_sha256":"1928fea93afe8fc26f867795f73730e315849176cc0c68f9a5d08f4875301ff5","source":{"kind":"arxiv","id":"2607.27933","version":1},"attestation_state":"computed","paper":{"title":"The Geometric Nature and a Free Proxy for Flow-Matching Uncertainty","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Ben Fei, Hui Xiong, Weiyu Guo, Yandong Guo, Yiren Zhao, Ziyang Rao","submitted_at":"2026-07-30T09:44:46Z","abstract_excerpt":"Flow matching (FM) has become a popular action head paradigm for modern embodied models. However, as a conditional generative model, it does not explicitly expose its inherent uncertainty, producing faulty action chunks even when it misinterprets the scene or encounters out-of-distribution (OOD) inputs. Therefore, determining when an FM-generated action can be trusted is essential for safe deployment, yet existing uncertainty estimation methods on real-time control suffer from several issues: extra training budget, high computational overhead, and low generalization ability. In this work, we p"},"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":"2607.27933","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.AI","submitted_at":"2026-07-30T09:44:46Z","cross_cats_sorted":[],"title_canon_sha256":"f61e5a73c7d318b141bd9f46fe23acb85b0c273613902399f223f188c5a00d72","abstract_canon_sha256":"fcb5fb9c2ae749eb225c4bf3366adf72190be37a4409b7ad418bb646dc5c61f8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1928fea93afe8fc26f867795f73730e315849176cc0c68f9a5d08f4875301ff5","last_reissued_at":"2026-07-31T01:34:50.096414Z","signature_status":"unsigned_v0","first_computed_at":"2026-07-31T01:34:50.096414Z"},"graph_snapshot":{"paper":{"title":"The Geometric Nature and a Free Proxy for Flow-Matching Uncertainty","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.AI","authors_text":"Ben Fei, Hui Xiong, Weiyu Guo, Yandong Guo, Yiren Zhao, Ziyang Rao","submitted_at":"2026-07-30T09:44:46Z","abstract_excerpt":"Flow matching (FM) has become a popular action head paradigm for modern embodied models. However, as a conditional generative model, it does not explicitly expose its inherent uncertainty, producing faulty action chunks even when it misinterprets the scene or encounters out-of-distribution (OOD) inputs. Therefore, determining when an FM-generated action can be trusted is essential for safe deployment, yet existing uncertainty estimation methods on real-time control suffer from several issues: extra training budget, high computational overhead, and low generalization ability. In this work, we p"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.27933","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/2607.27933/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":"2607.27933","created_at":"2026-07-31T01:34:50.100128+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.27933v1","created_at":"2026-07-31T01:34:50.100128+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.27933","created_at":"2026-07-31T01:34:50.100128+00:00"},{"alias_kind":"pith_short_12","alias_value":"DEUP5KJ272H4","created_at":"2026-07-31T01:34:50.100128+00:00"},{"alias_kind":"pith_short_16","alias_value":"DEUP5KJ272H4E34G","created_at":"2026-07-31T01:34:50.100128+00:00"},{"alias_kind":"pith_short_8","alias_value":"DEUP5KJ2","created_at":"2026-07-31T01:34:50.100128+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2608.11521","citing_title":"Keep the Future, Drop the Rollout: RIFT for World Action Models","ref_index":55,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/DEUP5KJ272H4E34GO6K7ONZQ4M","json":"https://pith.science/pith/DEUP5KJ272H4E34GO6K7ONZQ4M.json","graph_json":"https://pith.science/api/pith-number/DEUP5KJ272H4E34GO6K7ONZQ4M/graph.json","events_json":"https://pith.science/api/pith-number/DEUP5KJ272H4E34GO6K7ONZQ4M/events.json","paper":"https://pith.science/paper/DEUP5KJ2"},"agent_actions":{"view_html":"https://pith.science/pith/DEUP5KJ272H4E34GO6K7ONZQ4M","download_json":"https://pith.science/pith/DEUP5KJ272H4E34GO6K7ONZQ4M.json","view_paper":"https://pith.science/paper/DEUP5KJ2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.27933&json=true","fetch_graph":"https://pith.science/api/pith-number/DEUP5KJ272H4E34GO6K7ONZQ4M/graph.json","fetch_events":"https://pith.science/api/pith-number/DEUP5KJ272H4E34GO6K7ONZQ4M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DEUP5KJ272H4E34GO6K7ONZQ4M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DEUP5KJ272H4E34GO6K7ONZQ4M/action/storage_attestation","attest_author":"https://pith.science/pith/DEUP5KJ272H4E34GO6K7ONZQ4M/action/author_attestation","sign_citation":"https://pith.science/pith/DEUP5KJ272H4E34GO6K7ONZQ4M/action/citation_signature","submit_replication":"https://pith.science/pith/DEUP5KJ272H4E34GO6K7ONZQ4M/action/replication_record"}},"created_at":"2026-07-31T01:34:50.100128+00:00","updated_at":"2026-07-31T01:34:50.100128+00:00"}