{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:62G5EWQ5U4QFQT6GCVIUCWKP5K","short_pith_number":"pith:62G5EWQ5","schema_version":"1.0","canonical_sha256":"f68dd25a1da720584fc6155141594fea9a7b806809b06017ce7a48a6f455d1df","source":{"kind":"arxiv","id":"2606.00773","version":1},"attestation_state":"computed","paper":{"title":"SafeVLA-Bench: A Benchmark for the Success-Safety Gap in Vision-Language-Action Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Fanxin Kong, Insup Lee, Jialiang Fan, Oleg Sokolsky, Weizhe Xu","submitted_at":"2026-05-30T15:31:57Z","abstract_excerpt":"Vision-language-action (VLA) benchmarks measure whether a policy completes a requested manipulation task, but binary success can hide safety-relevant trajectory behavior: reaching the goal while applying excessive contact, disturbing bystander objects, destabilizing the held object, or entering robot self-contact. We present SafeVLA-Bench, a post-hoc safety-evaluation framework for existing simulator-based VLA benchmarks. It formalizes task-aware safety requirements as Signal Temporal Logic (STL) specifications and reports native success with two unsafe-success metrics: Succ-But-Unsafe (SBU), "},"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":"2606.00773","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.RO","submitted_at":"2026-05-30T15:31:57Z","cross_cats_sorted":[],"title_canon_sha256":"d71bc79e4f9cbcce223a1d2f92b48759c24278eb2569c27ee71e280f29598556","abstract_canon_sha256":"5b80746f8faa37627656cff318c9156301b29a91f4a941dde6452974f01af626"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-02T01:04:05.363003Z","signature_b64":"b/NzB86jsJMLngash8kP+YrrYxY8KasO8x74IzVTuUUpzdL9Gkbq+RmN9OdTxCdbeqtxT/46hsMfUCUM2YHlCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f68dd25a1da720584fc6155141594fea9a7b806809b06017ce7a48a6f455d1df","last_reissued_at":"2026-06-02T01:04:05.362590Z","signature_status":"signed_v1","first_computed_at":"2026-06-02T01:04:05.362590Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"SafeVLA-Bench: A Benchmark for the Success-Safety Gap in Vision-Language-Action Models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.RO","authors_text":"Fanxin Kong, Insup Lee, Jialiang Fan, Oleg Sokolsky, Weizhe Xu","submitted_at":"2026-05-30T15:31:57Z","abstract_excerpt":"Vision-language-action (VLA) benchmarks measure whether a policy completes a requested manipulation task, but binary success can hide safety-relevant trajectory behavior: reaching the goal while applying excessive contact, disturbing bystander objects, destabilizing the held object, or entering robot self-contact. We present SafeVLA-Bench, a post-hoc safety-evaluation framework for existing simulator-based VLA benchmarks. It formalizes task-aware safety requirements as Signal Temporal Logic (STL) specifications and reports native success with two unsafe-success metrics: Succ-But-Unsafe (SBU), "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.00773","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/2606.00773/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":"2606.00773","created_at":"2026-06-02T01:04:05.362650+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.00773v1","created_at":"2026-06-02T01:04:05.362650+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.00773","created_at":"2026-06-02T01:04:05.362650+00:00"},{"alias_kind":"pith_short_12","alias_value":"62G5EWQ5U4QF","created_at":"2026-06-02T01:04:05.362650+00:00"},{"alias_kind":"pith_short_16","alias_value":"62G5EWQ5U4QFQT6G","created_at":"2026-06-02T01:04:05.362650+00:00"},{"alias_kind":"pith_short_8","alias_value":"62G5EWQ5","created_at":"2026-06-02T01:04:05.362650+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/62G5EWQ5U4QFQT6GCVIUCWKP5K","json":"https://pith.science/pith/62G5EWQ5U4QFQT6GCVIUCWKP5K.json","graph_json":"https://pith.science/api/pith-number/62G5EWQ5U4QFQT6GCVIUCWKP5K/graph.json","events_json":"https://pith.science/api/pith-number/62G5EWQ5U4QFQT6GCVIUCWKP5K/events.json","paper":"https://pith.science/paper/62G5EWQ5"},"agent_actions":{"view_html":"https://pith.science/pith/62G5EWQ5U4QFQT6GCVIUCWKP5K","download_json":"https://pith.science/pith/62G5EWQ5U4QFQT6GCVIUCWKP5K.json","view_paper":"https://pith.science/paper/62G5EWQ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.00773&json=true","fetch_graph":"https://pith.science/api/pith-number/62G5EWQ5U4QFQT6GCVIUCWKP5K/graph.json","fetch_events":"https://pith.science/api/pith-number/62G5EWQ5U4QFQT6GCVIUCWKP5K/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/62G5EWQ5U4QFQT6GCVIUCWKP5K/action/timestamp_anchor","attest_storage":"https://pith.science/pith/62G5EWQ5U4QFQT6GCVIUCWKP5K/action/storage_attestation","attest_author":"https://pith.science/pith/62G5EWQ5U4QFQT6GCVIUCWKP5K/action/author_attestation","sign_citation":"https://pith.science/pith/62G5EWQ5U4QFQT6GCVIUCWKP5K/action/citation_signature","submit_replication":"https://pith.science/pith/62G5EWQ5U4QFQT6GCVIUCWKP5K/action/replication_record"}},"created_at":"2026-06-02T01:04:05.362650+00:00","updated_at":"2026-06-02T01:04:05.362650+00:00"}