{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:56GWHSSX22L4FBE6S7D3WLCK2W","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e48b1d6e7014fefe724fb74a1bce032a262a5f70c8fb634c1c9da864a8b24a10","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-05-12T16:49:28Z","title_canon_sha256":"c33e6927696d000caf790804a5cdb6c955cff6d8f248b7bd48ea53f3c7078a90"},"schema_version":"1.0","source":{"id":"2605.12386","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.12386","created_at":"2026-06-11T02:09:30Z"},{"alias_kind":"arxiv_version","alias_value":"2605.12386v2","created_at":"2026-06-11T02:09:30Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.12386","created_at":"2026-06-11T02:09:30Z"},{"alias_kind":"pith_short_12","alias_value":"56GWHSSX22L4","created_at":"2026-06-11T02:09:30Z"},{"alias_kind":"pith_short_16","alias_value":"56GWHSSX22L4FBE6","created_at":"2026-06-11T02:09:30Z"},{"alias_kind":"pith_short_8","alias_value":"56GWHSSX","created_at":"2026-06-11T02:09:30Z"}],"graph_snapshots":[{"event_id":"sha256:4df6965560a25980f0028d441dac99d208ada99dc3e078a4d0cc3f2c5e1b7463","target":"graph","created_at":"2026-06-11T02:09:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Results show that even strong models often behave unsafely. Task-success gains do not reliably translate into safer execution: many successful rollouts remain unsafe, while longer-horizon or more complex tasks expose more violations."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"The LTLf safety templates and the mapping from observed rollouts to symbolic predicate traces accurately capture all relevant temporal safety properties without introducing false positives or missing critical real-world constraints."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"SafeManip is a new benchmark that applies LTLf monitors to assess temporal safety properties across eight categories in robotic manipulation, demonstrating that task success frequently fails to ensure safe execution in vision-language-action policies."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"A new benchmark shows that robotic manipulation policies often violate temporal safety rules even on tasks they complete successfully."}],"snapshot_sha256":"0a1876392ab6438af818d0cf487a218824735b37fcc5b8a4363a41c00fd095a0"},"formal_canon":{"evidence_count":2,"snapshot_sha256":"224f6469a73e6cb72e5afe6c57d5a00029b466606582210e56e7e21df4e900bc"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-26T14:40:19.643466Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_title_agreement","ran_at":"2026-05-20T13:31:25.053122Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-20T09:46:22.567775Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"claim_evidence","ran_at":"2026-05-19T22:41:58.221082Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2605.12386/integrity.json","findings":[],"snapshot_sha256":"5887839b98fc958103d3559f742d458d8d827d2aa36504639bc88d0cb4f5c9e2","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Robotic manipulation is typically evaluated by task success, but successful completion does not guarantee safe execution. Many safety failures are temporal: a robot may touch a clean surface after contamination or release an object before it is fully inside an enclosure. We introduce SafeManip, a property-driven benchmark to explicitly evaluate temporal safety properties in robotic manipulation, moving beyond prior evaluations that largely focus on task completion or per-state constraint violations. SafeManip defines reusable safety templates over finite executions using Linear Temporal Logic ","authors_text":"Chengyue Huang, Khang Vo Huynh, Lu Feng, Sebastian Elbaum, Zsolt Kira","cross_cats":[],"headline":"A new benchmark shows that robotic manipulation policies often violate temporal safety rules even on tasks they complete successfully.","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-05-12T16:49:28Z","title":"SafeManip: A Property-Driven Benchmark for Temporal Safety Evaluation in Robotic Manipulation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.12386","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-13T03:36:05.803212Z","id":"850d389d-b100-40ee-8ba1-39779118c924","model_set":{"reader":"grok-4.3"},"one_line_summary":"SafeManip is a new benchmark that applies LTLf monitors to assess temporal safety properties across eight categories in robotic manipulation, demonstrating that task success frequently fails to ensure safe execution in vision-language-action policies.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"A new benchmark shows that robotic manipulation policies often violate temporal safety rules even on tasks they complete successfully.","strongest_claim":"Results show that even strong models often behave unsafely. Task-success gains do not reliably translate into safer execution: many successful rollouts remain unsafe, while longer-horizon or more complex tasks expose more violations.","weakest_assumption":"The LTLf safety templates and the mapping from observed rollouts to symbolic predicate traces accurately capture all relevant temporal safety properties without introducing false positives or missing critical real-world constraints."}},"verdict_id":"850d389d-b100-40ee-8ba1-39779118c924"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:9d1cb095e3186f2e744200e71eced0dbd0473cb1c860fd5aaff9862cf4faf07f","target":"record","created_at":"2026-06-11T02:09:30Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e48b1d6e7014fefe724fb74a1bce032a262a5f70c8fb634c1c9da864a8b24a10","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.RO","submitted_at":"2026-05-12T16:49:28Z","title_canon_sha256":"c33e6927696d000caf790804a5cdb6c955cff6d8f248b7bd48ea53f3c7078a90"},"schema_version":"1.0","source":{"id":"2605.12386","kind":"arxiv","version":2}},"canonical_sha256":"ef8d63ca57d697c2849e97c7bb2c4ad5be7da783abc59f60446a80065f37b7c0","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ef8d63ca57d697c2849e97c7bb2c4ad5be7da783abc59f60446a80065f37b7c0","first_computed_at":"2026-06-11T02:09:30.756267Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-11T02:09:30.756267Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"gFhOLu1lIB7IicOmvjuHaI6erPoe1sM8xlPQKefx9FmgsUUWzI0Wzw6duR9vF6M9mcXkhyBImMTkO/CqwkyYBQ==","signature_status":"signed_v1","signed_at":"2026-06-11T02:09:30.757517Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.12386","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9d1cb095e3186f2e744200e71eced0dbd0473cb1c860fd5aaff9862cf4faf07f","sha256:4df6965560a25980f0028d441dac99d208ada99dc3e078a4d0cc3f2c5e1b7463"],"state_sha256":"73f37f81e6e684ef4b00a797ca3a26cde58c3eb8ccccd9f2172b0f54f05e4b7b"}