{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:FMS25BOZTMZL6EE7ESHI2M4I7M","short_pith_number":"pith:FMS25BOZ","schema_version":"1.0","canonical_sha256":"2b25ae85d99b32bf109f248e8d3388fb22e54abffbf296aa3154643f945851c2","source":{"kind":"arxiv","id":"2410.01971","version":1},"attestation_state":"computed","paper":{"title":"Run-time Observation Interventions Make Vision-Language-Action Models More Visually Robust","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Allen Z. Ren, Anirudha Majumdar, Asher J. Hancock","submitted_at":"2024-10-02T19:29:24Z","abstract_excerpt":"Vision-language-action (VLA) models trained on large-scale internet data and robot demonstrations have the potential to serve as generalist robot policies. However, despite their large-scale training, VLAs are often brittle to task-irrelevant visual details such as distractor objects or background colors. We introduce Bring Your Own VLA (BYOVLA): a run-time intervention scheme that (1) dynamically identifies regions of the input image that the model is sensitive to, and (2) minimally alters task-irrelevant regions to reduce the model's sensitivity using automated image editing tools. Our appro"},"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":"2410.01971","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2024-10-02T19:29:24Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"03ba5556863b6198dfd8ce4737bb09d222bb473c4f89216c7544d43e06eab9af","abstract_canon_sha256":"c0788ab0d3567e8c7dedcce3913ad3c1df4049c5a19d5ec72df32bc886132a98"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:14:58.617604Z","signature_b64":"vrUleAH6+5nZGQo9MVgZOvLgKnDW8HFwa+dk4DGV+y1TnXPd63zBDMXP9NdY0vPOtzwdSrgMsRGc0bFvn8ixAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2b25ae85d99b32bf109f248e8d3388fb22e54abffbf296aa3154643f945851c2","last_reissued_at":"2026-07-05T09:14:58.617178Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:14:58.617178Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Run-time Observation Interventions Make Vision-Language-Action Models More Visually Robust","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.RO","authors_text":"Allen Z. Ren, Anirudha Majumdar, Asher J. Hancock","submitted_at":"2024-10-02T19:29:24Z","abstract_excerpt":"Vision-language-action (VLA) models trained on large-scale internet data and robot demonstrations have the potential to serve as generalist robot policies. However, despite their large-scale training, VLAs are often brittle to task-irrelevant visual details such as distractor objects or background colors. We introduce Bring Your Own VLA (BYOVLA): a run-time intervention scheme that (1) dynamically identifies regions of the input image that the model is sensitive to, and (2) minimally alters task-irrelevant regions to reduce the model's sensitivity using automated image editing tools. Our appro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2410.01971","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/2410.01971/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":"2410.01971","created_at":"2026-07-05T09:14:58.617235+00:00"},{"alias_kind":"arxiv_version","alias_value":"2410.01971v1","created_at":"2026-07-05T09:14:58.617235+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2410.01971","created_at":"2026-07-05T09:14:58.617235+00:00"},{"alias_kind":"pith_short_12","alias_value":"FMS25BOZTMZL","created_at":"2026-07-05T09:14:58.617235+00:00"},{"alias_kind":"pith_short_16","alias_value":"FMS25BOZTMZL6EE7","created_at":"2026-07-05T09:14:58.617235+00:00"},{"alias_kind":"pith_short_8","alias_value":"FMS25BOZ","created_at":"2026-07-05T09:14:58.617235+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2508.10399","citing_title":"Large Model Empowered Embodied AI: A Survey on Decision-Making and Embodied Learning","ref_index":74,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FMS25BOZTMZL6EE7ESHI2M4I7M","json":"https://pith.science/pith/FMS25BOZTMZL6EE7ESHI2M4I7M.json","graph_json":"https://pith.science/api/pith-number/FMS25BOZTMZL6EE7ESHI2M4I7M/graph.json","events_json":"https://pith.science/api/pith-number/FMS25BOZTMZL6EE7ESHI2M4I7M/events.json","paper":"https://pith.science/paper/FMS25BOZ"},"agent_actions":{"view_html":"https://pith.science/pith/FMS25BOZTMZL6EE7ESHI2M4I7M","download_json":"https://pith.science/pith/FMS25BOZTMZL6EE7ESHI2M4I7M.json","view_paper":"https://pith.science/paper/FMS25BOZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2410.01971&json=true","fetch_graph":"https://pith.science/api/pith-number/FMS25BOZTMZL6EE7ESHI2M4I7M/graph.json","fetch_events":"https://pith.science/api/pith-number/FMS25BOZTMZL6EE7ESHI2M4I7M/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FMS25BOZTMZL6EE7ESHI2M4I7M/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FMS25BOZTMZL6EE7ESHI2M4I7M/action/storage_attestation","attest_author":"https://pith.science/pith/FMS25BOZTMZL6EE7ESHI2M4I7M/action/author_attestation","sign_citation":"https://pith.science/pith/FMS25BOZTMZL6EE7ESHI2M4I7M/action/citation_signature","submit_replication":"https://pith.science/pith/FMS25BOZTMZL6EE7ESHI2M4I7M/action/replication_record"}},"created_at":"2026-07-05T09:14:58.617235+00:00","updated_at":"2026-07-05T09:14:58.617235+00:00"}