{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:CIUIRQITKQXAFZJV7AC6UYJ3DP","short_pith_number":"pith:CIUIRQIT","canonical_record":{"source":{"id":"2402.04210","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-02-06T18:07:43Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"17d44da16829365f2f8fc03e0686dddd0685004edbff14814ec16e0dce80f2ed","abstract_canon_sha256":"44b851a5247e7eaff29c0cfb4e935bd096e95f4e5cfd170f590cda46e9c5e728"},"schema_version":"1.0"},"canonical_sha256":"122888c113542e02e535f805ea613b1bcc343ed5efa7d7ce19d793a7c8f033f3","source":{"kind":"arxiv","id":"2402.04210","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.04210","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"arxiv_version","alias_value":"2402.04210v2","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.04210","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"pith_short_12","alias_value":"CIUIRQITKQXA","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"pith_short_16","alias_value":"CIUIRQITKQXAFZJV","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"pith_short_8","alias_value":"CIUIRQIT","created_at":"2026-07-05T08:54:06Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:CIUIRQITKQXAFZJV7AC6UYJ3DP","target":"record","payload":{"canonical_record":{"source":{"id":"2402.04210","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-02-06T18:07:43Z","cross_cats_sorted":["cs.RO"],"title_canon_sha256":"17d44da16829365f2f8fc03e0686dddd0685004edbff14814ec16e0dce80f2ed","abstract_canon_sha256":"44b851a5247e7eaff29c0cfb4e935bd096e95f4e5cfd170f590cda46e9c5e728"},"schema_version":"1.0"},"canonical_sha256":"122888c113542e02e535f805ea613b1bcc343ed5efa7d7ce19d793a7c8f033f3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:54:06.997822Z","signature_b64":"zQ4Eg8KPJwDE/6T6UoaM6bD619mAWF/ATzp5IZWQfzpLKVfe5XBcUKXwHSeUN3Zgy1uMgxwT1NNasBvyxRWjDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"122888c113542e02e535f805ea613b1bcc343ed5efa7d7ce19d793a7c8f033f3","last_reissued_at":"2026-07-05T08:54:06.997354Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:54:06.997354Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2402.04210","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:54:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"zZkTe1qSXS+FkjqqNEKI9H8ZtIXkuxgYCJH0y5Je5Ge9wJGfGua10/Trgdmpis4fxhQ6LTOlqW3dZrTkgg/PDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T15:57:44.370491Z"},"content_sha256":"8aaace787a4c11d3b1c48a7346273f5bf77cbb6f2665809dde4e8027587954e9","schema_version":"1.0","event_id":"sha256:8aaace787a4c11d3b1c48a7346273f5bf77cbb6f2665809dde4e8027587954e9"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:CIUIRQITKQXAFZJV7AC6UYJ3DP","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Task Success is not Enough: Investigating the Use of Video-Language Models as Behavior Critics for Catching Undesirable Agent Behaviors","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.RO"],"primary_cat":"cs.AI","authors_text":"Denis Liu, Heni Ben Amor, Lin Guan, Subbarao Kambhampati, Yantian Zha, Yifan Zhou","submitted_at":"2024-02-06T18:07:43Z","abstract_excerpt":"Large-scale generative models are shown to be useful for sampling meaningful candidate solutions, yet they often overlook task constraints and user preferences. Their full power is better harnessed when the models are coupled with external verifiers and the final solutions are derived iteratively or progressively according to the verification feedback. In the context of embodied AI, verification often solely involves assessing whether goal conditions specified in the instructions have been met. Nonetheless, for these agents to be seamlessly integrated into daily life, it is crucial to account "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.04210","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/2402.04210/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:54:06Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"W7Acgrv/rfXyACDbVm2biMqVN8Pp4TYKKs3hSR0eTSpT2V6wvCAafGsfOBRu6llnTEdYMGHsMu1+ln7OjyteAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T15:57:44.371207Z"},"content_sha256":"daa60b6c86e0467747946fa2f8b2b91e92e3fd36b4a72c25fe60f4cd0f9b65ab","schema_version":"1.0","event_id":"sha256:daa60b6c86e0467747946fa2f8b2b91e92e3fd36b4a72c25fe60f4cd0f9b65ab"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/CIUIRQITKQXAFZJV7AC6UYJ3DP/bundle.json","state_url":"https://pith.science/pith/CIUIRQITKQXAFZJV7AC6UYJ3DP/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/CIUIRQITKQXAFZJV7AC6UYJ3DP/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-05T15:57:44Z","links":{"resolver":"https://pith.science/pith/CIUIRQITKQXAFZJV7AC6UYJ3DP","bundle":"https://pith.science/pith/CIUIRQITKQXAFZJV7AC6UYJ3DP/bundle.json","state":"https://pith.science/pith/CIUIRQITKQXAFZJV7AC6UYJ3DP/state.json","well_known_bundle":"https://pith.science/.well-known/pith/CIUIRQITKQXAFZJV7AC6UYJ3DP/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:CIUIRQITKQXAFZJV7AC6UYJ3DP","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":"44b851a5247e7eaff29c0cfb4e935bd096e95f4e5cfd170f590cda46e9c5e728","cross_cats_sorted":["cs.RO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-02-06T18:07:43Z","title_canon_sha256":"17d44da16829365f2f8fc03e0686dddd0685004edbff14814ec16e0dce80f2ed"},"schema_version":"1.0","source":{"id":"2402.04210","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2402.04210","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"arxiv_version","alias_value":"2402.04210v2","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2402.04210","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"pith_short_12","alias_value":"CIUIRQITKQXA","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"pith_short_16","alias_value":"CIUIRQITKQXAFZJV","created_at":"2026-07-05T08:54:06Z"},{"alias_kind":"pith_short_8","alias_value":"CIUIRQIT","created_at":"2026-07-05T08:54:06Z"}],"graph_snapshots":[{"event_id":"sha256:daa60b6c86e0467747946fa2f8b2b91e92e3fd36b4a72c25fe60f4cd0f9b65ab","target":"graph","created_at":"2026-07-05T08:54:06Z","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":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2402.04210/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Large-scale generative models are shown to be useful for sampling meaningful candidate solutions, yet they often overlook task constraints and user preferences. Their full power is better harnessed when the models are coupled with external verifiers and the final solutions are derived iteratively or progressively according to the verification feedback. In the context of embodied AI, verification often solely involves assessing whether goal conditions specified in the instructions have been met. Nonetheless, for these agents to be seamlessly integrated into daily life, it is crucial to account ","authors_text":"Denis Liu, Heni Ben Amor, Lin Guan, Subbarao Kambhampati, Yantian Zha, Yifan Zhou","cross_cats":["cs.RO"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-02-06T18:07:43Z","title":"Task Success is not Enough: Investigating the Use of Video-Language Models as Behavior Critics for Catching Undesirable Agent Behaviors"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2402.04210","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:8aaace787a4c11d3b1c48a7346273f5bf77cbb6f2665809dde4e8027587954e9","target":"record","created_at":"2026-07-05T08:54:06Z","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":"44b851a5247e7eaff29c0cfb4e935bd096e95f4e5cfd170f590cda46e9c5e728","cross_cats_sorted":["cs.RO"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2024-02-06T18:07:43Z","title_canon_sha256":"17d44da16829365f2f8fc03e0686dddd0685004edbff14814ec16e0dce80f2ed"},"schema_version":"1.0","source":{"id":"2402.04210","kind":"arxiv","version":2}},"canonical_sha256":"122888c113542e02e535f805ea613b1bcc343ed5efa7d7ce19d793a7c8f033f3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"122888c113542e02e535f805ea613b1bcc343ed5efa7d7ce19d793a7c8f033f3","first_computed_at":"2026-07-05T08:54:06.997354Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:54:06.997354Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"zQ4Eg8KPJwDE/6T6UoaM6bD619mAWF/ATzp5IZWQfzpLKVfe5XBcUKXwHSeUN3Zgy1uMgxwT1NNasBvyxRWjDw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:54:06.997822Z","signed_message":"canonical_sha256_bytes"},"source_id":"2402.04210","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:8aaace787a4c11d3b1c48a7346273f5bf77cbb6f2665809dde4e8027587954e9","sha256:daa60b6c86e0467747946fa2f8b2b91e92e3fd36b4a72c25fe60f4cd0f9b65ab"],"state_sha256":"83a79d30a3f41f210bb402f17aaa24302e91385bcec5e3237f6c43941ea2802a"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"scqCQZ0m4nDdhsljxJdJUIJSWgKSBmfwEKN2PNP0SI+bMawOt8dJVifbDbnCQPvBkTJ7BGYg5Ih8rnI+UQYyCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-05T15:57:44.375188Z","bundle_sha256":"24646422e852acbf0302eabd8ce01a019a64d7d45b432bc76634d3c9d18931ce"}}