{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XDLO6YQZL3B3POEOFSIYSYNSEG","short_pith_number":"pith:XDLO6YQZ","schema_version":"1.0","canonical_sha256":"b8d6ef62195ec3b7b88e2c918961b221817ce993a97423d008a4e6d0fc4bfd42","source":{"kind":"arxiv","id":"2510.01389","version":2},"attestation_state":"computed","paper":{"title":"INSIGHT: INference-time Sequence Introspection for Generating Help Triggers in Vision-Language-Action Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Tesca FItzgerald, Ulas Berk Karli, Ziyao Shangguan","submitted_at":"2025-10-01T19:22:48Z","abstract_excerpt":"Recent Vision-Language-Action (VLA) models show strong generalization capabilities, yet they lack introspective mechanisms for anticipating failures and requesting help from a human supervisor. We present \\textbf{INSIGHT}, a learning framework for leveraging token-level uncertainty signals to predict when a VLA should request help. Using $\\pi_0$-FAST as the underlying model, we extract per-token \\emph{entropy}, \\emph{log-probability}, and Dirichlet-based estimates of \\emph{aleatoric and epistemic uncertainty}, and train compact transformer classifiers to map these sequences to help triggers. W"},"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":"2510.01389","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.RO","submitted_at":"2025-10-01T19:22:48Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"5bd1f4750af19e8723ea4b46db7bc1fbfbbb3f9f634718b8d8fe17551915b8e2","abstract_canon_sha256":"f460a9b447dfadd118c632821a6ebb0a152f931422690f6e69b1df29d63523e7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T02:03:56.314582Z","signature_b64":"J/gTh1ve1j2OEV4L5p5CnoDim55EP6IVvJI/Ev6PbrqdLjdh2/7X4BWGPdw1Mr6fjTehn5ivOSvaeBwBn5s4AA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b8d6ef62195ec3b7b88e2c918961b221817ce993a97423d008a4e6d0fc4bfd42","last_reissued_at":"2026-05-26T02:03:56.313693Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T02:03:56.313693Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"INSIGHT: INference-time Sequence Introspection for Generating Help Triggers in Vision-Language-Action Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.RO","authors_text":"Tesca FItzgerald, Ulas Berk Karli, Ziyao Shangguan","submitted_at":"2025-10-01T19:22:48Z","abstract_excerpt":"Recent Vision-Language-Action (VLA) models show strong generalization capabilities, yet they lack introspective mechanisms for anticipating failures and requesting help from a human supervisor. We present \\textbf{INSIGHT}, a learning framework for leveraging token-level uncertainty signals to predict when a VLA should request help. Using $\\pi_0$-FAST as the underlying model, we extract per-token \\emph{entropy}, \\emph{log-probability}, and Dirichlet-based estimates of \\emph{aleatoric and epistemic uncertainty}, and train compact transformer classifiers to map these sequences to help triggers. W"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2510.01389","kind":"arxiv","version":2},"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/2510.01389/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":"2510.01389","created_at":"2026-05-26T02:03:56.313824+00:00"},{"alias_kind":"arxiv_version","alias_value":"2510.01389v2","created_at":"2026-05-26T02:03:56.313824+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2510.01389","created_at":"2026-05-26T02:03:56.313824+00:00"},{"alias_kind":"pith_short_12","alias_value":"XDLO6YQZL3B3","created_at":"2026-05-26T02:03:56.313824+00:00"},{"alias_kind":"pith_short_16","alias_value":"XDLO6YQZL3B3POEO","created_at":"2026-05-26T02:03:56.313824+00:00"},{"alias_kind":"pith_short_8","alias_value":"XDLO6YQZ","created_at":"2026-05-26T02:03:56.313824+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/XDLO6YQZL3B3POEOFSIYSYNSEG","json":"https://pith.science/pith/XDLO6YQZL3B3POEOFSIYSYNSEG.json","graph_json":"https://pith.science/api/pith-number/XDLO6YQZL3B3POEOFSIYSYNSEG/graph.json","events_json":"https://pith.science/api/pith-number/XDLO6YQZL3B3POEOFSIYSYNSEG/events.json","paper":"https://pith.science/paper/XDLO6YQZ"},"agent_actions":{"view_html":"https://pith.science/pith/XDLO6YQZL3B3POEOFSIYSYNSEG","download_json":"https://pith.science/pith/XDLO6YQZL3B3POEOFSIYSYNSEG.json","view_paper":"https://pith.science/paper/XDLO6YQZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2510.01389&json=true","fetch_graph":"https://pith.science/api/pith-number/XDLO6YQZL3B3POEOFSIYSYNSEG/graph.json","fetch_events":"https://pith.science/api/pith-number/XDLO6YQZL3B3POEOFSIYSYNSEG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XDLO6YQZL3B3POEOFSIYSYNSEG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XDLO6YQZL3B3POEOFSIYSYNSEG/action/storage_attestation","attest_author":"https://pith.science/pith/XDLO6YQZL3B3POEOFSIYSYNSEG/action/author_attestation","sign_citation":"https://pith.science/pith/XDLO6YQZL3B3POEOFSIYSYNSEG/action/citation_signature","submit_replication":"https://pith.science/pith/XDLO6YQZL3B3POEOFSIYSYNSEG/action/replication_record"}},"created_at":"2026-05-26T02:03:56.313824+00:00","updated_at":"2026-05-26T02:03:56.313824+00:00"}