{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:EQRE24FC7HDNCE4AIVCRCFTJYR","short_pith_number":"pith:EQRE24FC","schema_version":"1.0","canonical_sha256":"24224d70a2f9c6d113804545111669c47778825c3f687d5238c1e6858ce8f70b","source":{"kind":"arxiv","id":"2503.08558","version":3},"attestation_state":"computed","paper":{"title":"Can We Detect Failures Without Failure Data? Uncertainty-Aware Runtime Failure Detection for Imitation Learning Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Chen Xu, Christopher Rodriguez, Emma Dixon, Haruki Nishimura, Masha Itkina, Paarth Shah, Patrick Miller, Rares Ambrus, Robert Lee, Tony Khuong Nguyen","submitted_at":"2025-03-11T15:47:12Z","abstract_excerpt":"Recent years have witnessed impressive robotic manipulation systems driven by advances in imitation learning and generative modeling, such as diffusion- and flow-based approaches. As robot policy performance increases, so does the complexity and time horizon of achievable tasks, inducing unexpected and diverse failure modes that are difficult to predict a priori. To enable trustworthy policy deployment in safety-critical human environments, reliable runtime failure detection becomes important during policy inference. However, most existing failure detection approaches rely on prior knowledge o"},"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":"2503.08558","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2025-03-11T15:47:12Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"9ed902b48adb335c83f1be5fd356befe9410a5f5dfd947a2831be231bd2c9798","abstract_canon_sha256":"f0c379cdbd515b4e12542f0f5ebb00318993900f18f9e782bc9343c8dd2a318a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:24:16.195819Z","signature_b64":"PV7d+a09HYZbCXTHHBiFGqL4zBmyAnlhuyxHIiHcsJjtfWs4hCdYpKUXe23M5lad6UPVtzkx45W+trDt3LVFDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"24224d70a2f9c6d113804545111669c47778825c3f687d5238c1e6858ce8f70b","last_reissued_at":"2026-07-05T11:24:16.195225Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:24:16.195225Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Can We Detect Failures Without Failure Data? Uncertainty-Aware Runtime Failure Detection for Imitation Learning Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Chen Xu, Christopher Rodriguez, Emma Dixon, Haruki Nishimura, Masha Itkina, Paarth Shah, Patrick Miller, Rares Ambrus, Robert Lee, Tony Khuong Nguyen","submitted_at":"2025-03-11T15:47:12Z","abstract_excerpt":"Recent years have witnessed impressive robotic manipulation systems driven by advances in imitation learning and generative modeling, such as diffusion- and flow-based approaches. As robot policy performance increases, so does the complexity and time horizon of achievable tasks, inducing unexpected and diverse failure modes that are difficult to predict a priori. To enable trustworthy policy deployment in safety-critical human environments, reliable runtime failure detection becomes important during policy inference. However, most existing failure detection approaches rely on prior knowledge o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.08558","kind":"arxiv","version":3},"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/2503.08558/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":"2503.08558","created_at":"2026-07-05T11:24:16.195334+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.08558v3","created_at":"2026-07-05T11:24:16.195334+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.08558","created_at":"2026-07-05T11:24:16.195334+00:00"},{"alias_kind":"pith_short_12","alias_value":"EQRE24FC7HDN","created_at":"2026-07-05T11:24:16.195334+00:00"},{"alias_kind":"pith_short_16","alias_value":"EQRE24FC7HDNCE4A","created_at":"2026-07-05T11:24:16.195334+00:00"},{"alias_kind":"pith_short_8","alias_value":"EQRE24FC","created_at":"2026-07-05T11:24:16.195334+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":9,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.23085","citing_title":"Foresight: Failure Detection for Long-Horizon Robotic Manipulation with Action-Conditioned World Model Latents","ref_index":1,"is_internal_anchor":false},{"citing_arxiv_id":"2606.21572","citing_title":"Robot Critics that Sweat the Small Stuff","ref_index":48,"is_internal_anchor":false},{"citing_arxiv_id":"2606.19998","citing_title":"Tri-Info: Generalizable, Interpretable Failure Prediction for VLA Models via Information Theory","ref_index":297,"is_internal_anchor":false},{"citing_arxiv_id":"2606.08508","citing_title":"ActProbe: Action-Space Probe for Early Failure Detection of Generative Robot Policies","ref_index":23,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06660","citing_title":"AEGIS: A Backup Reflex for Physical AI","ref_index":28,"is_internal_anchor":false},{"citing_arxiv_id":"2606.06461","citing_title":"Flow-based Policy Adaptation without Policy Updates","ref_index":4,"is_internal_anchor":false},{"citing_arxiv_id":"2605.29605","citing_title":"VLAConf: Calibrated Task-Success Confidence for Vision-Language-Action Models","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2510.24680","citing_title":"InFeR: Informed Failure Resilience in Learned Visual Navigation Control","ref_index":22,"is_internal_anchor":false},{"citing_arxiv_id":"2604.13788","citing_title":"Failure Identification in Imitation Learning Via Statistical and Semantic Filtering","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/EQRE24FC7HDNCE4AIVCRCFTJYR","json":"https://pith.science/pith/EQRE24FC7HDNCE4AIVCRCFTJYR.json","graph_json":"https://pith.science/api/pith-number/EQRE24FC7HDNCE4AIVCRCFTJYR/graph.json","events_json":"https://pith.science/api/pith-number/EQRE24FC7HDNCE4AIVCRCFTJYR/events.json","paper":"https://pith.science/paper/EQRE24FC"},"agent_actions":{"view_html":"https://pith.science/pith/EQRE24FC7HDNCE4AIVCRCFTJYR","download_json":"https://pith.science/pith/EQRE24FC7HDNCE4AIVCRCFTJYR.json","view_paper":"https://pith.science/paper/EQRE24FC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.08558&json=true","fetch_graph":"https://pith.science/api/pith-number/EQRE24FC7HDNCE4AIVCRCFTJYR/graph.json","fetch_events":"https://pith.science/api/pith-number/EQRE24FC7HDNCE4AIVCRCFTJYR/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/EQRE24FC7HDNCE4AIVCRCFTJYR/action/timestamp_anchor","attest_storage":"https://pith.science/pith/EQRE24FC7HDNCE4AIVCRCFTJYR/action/storage_attestation","attest_author":"https://pith.science/pith/EQRE24FC7HDNCE4AIVCRCFTJYR/action/author_attestation","sign_citation":"https://pith.science/pith/EQRE24FC7HDNCE4AIVCRCFTJYR/action/citation_signature","submit_replication":"https://pith.science/pith/EQRE24FC7HDNCE4AIVCRCFTJYR/action/replication_record"}},"created_at":"2026-07-05T11:24:16.195334+00:00","updated_at":"2026-07-05T11:24:16.195334+00:00"}