{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:XAZL74HQLHKUVME54UL3WHZX2K","short_pith_number":"pith:XAZL74HQ","schema_version":"1.0","canonical_sha256":"b832bff0f059d54ab09de517bb1f37d2807fa2757bd879c98a14d289d2b728af","source":{"kind":"arxiv","id":"2606.29699","version":1},"attestation_state":"computed","paper":{"title":"Early Warning Signals for OpenVLA Failure under Visual Distribution Shift","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Dipesh Tharu Mahato, Rachel Ren","submitted_at":"2026-06-29T02:07:17Z","abstract_excerpt":"Vision Language Action models combine perception, language grounding, and control in a single policy, but their failures are hard to diagnose once visual conditions shift. We test whether OpenVLA feedforward activations contain linearly decodable information about near term task failure in LIBERO manipulation rollouts. The policy is fixed throughout. We log internal activations during execution and fit lightweight monitors after the rollouts are collected. Occlusion is the main controlled stress test. It reduces OpenVLA success from $57\\%$ to $17\\%$ over $100$ episodes per condition. Under thi"},"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":"2606.29699","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2026-06-29T02:07:17Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"ecbd8dddc25688a49255e2fe1407bd36a436917c573433ea1f61e795227c2f3c","abstract_canon_sha256":"be3b8584e0d17f568d4f36ab4998da5bf92ac1a0f33406980ec06efb507360c6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-30T02:17:31.126656Z","signature_b64":"fR1yIkVwD73ASsQ93sVbKdAceJskFbb8HfVMkPVgmma0rCsqBq/RWyFvNhLKYj8AQ+yAQjb1rfpehtUu96eqCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b832bff0f059d54ab09de517bb1f37d2807fa2757bd879c98a14d289d2b728af","last_reissued_at":"2026-06-30T02:17:31.126167Z","signature_status":"signed_v1","first_computed_at":"2026-06-30T02:17:31.126167Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Early Warning Signals for OpenVLA Failure under Visual Distribution Shift","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Dipesh Tharu Mahato, Rachel Ren","submitted_at":"2026-06-29T02:07:17Z","abstract_excerpt":"Vision Language Action models combine perception, language grounding, and control in a single policy, but their failures are hard to diagnose once visual conditions shift. We test whether OpenVLA feedforward activations contain linearly decodable information about near term task failure in LIBERO manipulation rollouts. The policy is fixed throughout. We log internal activations during execution and fit lightweight monitors after the rollouts are collected. Occlusion is the main controlled stress test. It reduces OpenVLA success from $57\\%$ to $17\\%$ over $100$ episodes per condition. Under thi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.29699","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/2606.29699/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":"2606.29699","created_at":"2026-06-30T02:17:31.126235+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.29699v1","created_at":"2026-06-30T02:17:31.126235+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.29699","created_at":"2026-06-30T02:17:31.126235+00:00"},{"alias_kind":"pith_short_12","alias_value":"XAZL74HQLHKU","created_at":"2026-06-30T02:17:31.126235+00:00"},{"alias_kind":"pith_short_16","alias_value":"XAZL74HQLHKUVME5","created_at":"2026-06-30T02:17:31.126235+00:00"},{"alias_kind":"pith_short_8","alias_value":"XAZL74HQ","created_at":"2026-06-30T02:17:31.126235+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/XAZL74HQLHKUVME54UL3WHZX2K","json":"https://pith.science/pith/XAZL74HQLHKUVME54UL3WHZX2K.json","graph_json":"https://pith.science/api/pith-number/XAZL74HQLHKUVME54UL3WHZX2K/graph.json","events_json":"https://pith.science/api/pith-number/XAZL74HQLHKUVME54UL3WHZX2K/events.json","paper":"https://pith.science/paper/XAZL74HQ"},"agent_actions":{"view_html":"https://pith.science/pith/XAZL74HQLHKUVME54UL3WHZX2K","download_json":"https://pith.science/pith/XAZL74HQLHKUVME54UL3WHZX2K.json","view_paper":"https://pith.science/paper/XAZL74HQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.29699&json=true","fetch_graph":"https://pith.science/api/pith-number/XAZL74HQLHKUVME54UL3WHZX2K/graph.json","fetch_events":"https://pith.science/api/pith-number/XAZL74HQLHKUVME54UL3WHZX2K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XAZL74HQLHKUVME54UL3WHZX2K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XAZL74HQLHKUVME54UL3WHZX2K/action/storage_attestation","attest_author":"https://pith.science/pith/XAZL74HQLHKUVME54UL3WHZX2K/action/author_attestation","sign_citation":"https://pith.science/pith/XAZL74HQLHKUVME54UL3WHZX2K/action/citation_signature","submit_replication":"https://pith.science/pith/XAZL74HQLHKUVME54UL3WHZX2K/action/replication_record"}},"created_at":"2026-06-30T02:17:31.126235+00:00","updated_at":"2026-06-30T02:17:31.126235+00:00"}