{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:EDF5OQSH63YUYBTKOAEOXDCO2U","short_pith_number":"pith:EDF5OQSH","canonical_record":{"source":{"id":"2602.16666","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T18:05:44Z","cross_cats_sorted":["cs.CY","cs.LG"],"title_canon_sha256":"bb8e3624248be3d14e90779a8dee6272c67d4e88957368d19422585411373f18","abstract_canon_sha256":"68bddfce41a46d60f9c9ae6cb67140f556dee2e33f81a367a2188b9494131cde"},"schema_version":"1.0"},"canonical_sha256":"20cbd74247f6f14c066a7008eb8c4ed507935363e3ad8531ecb11e67e65c56dd","source":{"kind":"arxiv","id":"2602.16666","version":3},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.16666","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"arxiv_version","alias_value":"2602.16666v3","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.16666","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"pith_short_12","alias_value":"EDF5OQSH63YU","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"pith_short_16","alias_value":"EDF5OQSH63YUYBTK","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"pith_short_8","alias_value":"EDF5OQSH","created_at":"2026-06-03T01:05:10Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:EDF5OQSH63YUYBTKOAEOXDCO2U","target":"record","payload":{"canonical_record":{"source":{"id":"2602.16666","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T18:05:44Z","cross_cats_sorted":["cs.CY","cs.LG"],"title_canon_sha256":"bb8e3624248be3d14e90779a8dee6272c67d4e88957368d19422585411373f18","abstract_canon_sha256":"68bddfce41a46d60f9c9ae6cb67140f556dee2e33f81a367a2188b9494131cde"},"schema_version":"1.0"},"canonical_sha256":"20cbd74247f6f14c066a7008eb8c4ed507935363e3ad8531ecb11e67e65c56dd","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-03T01:05:10.783503Z","signature_b64":"mbNtPIidtxqBzBepFF4EB8QpNFFZR/37Me4une6AIJ8Y5bY/QaYyX+aFGkdVjNnRji7YgLdRe8q+BEmDmMxwBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"20cbd74247f6f14c066a7008eb8c4ed507935363e3ad8531ecb11e67e65c56dd","last_reissued_at":"2026-06-03T01:05:10.783069Z","signature_status":"signed_v1","first_computed_at":"2026-06-03T01:05:10.783069Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2602.16666","source_version":3,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-03T01:05:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3aEYtO0y2LjbWdew4RuMOVf3po0rKovzlUcFJCUsyTyeutV46SBVegfV+4mP7LrMJbcXrNYHMaRclZIpsJrpAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T04:31:23.964966Z"},"content_sha256":"7e1cf77e22feab8e3c47b00dadbfbd7360fe7f06c3ff14127c7aa567243bd26d","schema_version":"1.0","event_id":"sha256:7e1cf77e22feab8e3c47b00dadbfbd7360fe7f06c3ff14127c7aa567243bd26d"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:EDF5OQSH63YUYBTKOAEOXDCO2U","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Towards a Science of AI Agent Reliability","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CY","cs.LG"],"primary_cat":"cs.AI","authors_text":"Arvind Narayanan, Kangheng Liu, Peter Kirgis, Saiteja Utpala, Sayash Kapoor, Stephan Rabanser","submitted_at":"2026-02-18T18:05:44Z","abstract_excerpt":"AI agents are increasingly deployed to execute important tasks. While rising accuracy scores on standard benchmarks suggest rapid progress, many agents still continue to fail in practice. This discrepancy highlights a fundamental limitation of current evaluations: compressing agent behavior into a single success metric obscures critical operational flaws. Notably, it ignores whether agents behave consistently across runs, withstand perturbations, fail predictably, or have bounded error severity. Grounded in safety-critical engineering, we provide a holistic performance profile by proposing twe"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.16666","kind":"arxiv","version":3},"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/2602.16666/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":"2026-06-03T01:05:10Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"u8+LEkIXSIIjWZTjVPGrS5k1jDcjg2IsRRE0iEzPdTLEy92Rig7JMmVKfPYVSWqRj/2MH68JoA8zWcI9cEyxDA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T04:31:23.965393Z"},"content_sha256":"6fc952d54f362765a50b5b63e2169eed246552d28ef24d1b735909c47204c7ba","schema_version":"1.0","event_id":"sha256:6fc952d54f362765a50b5b63e2169eed246552d28ef24d1b735909c47204c7ba"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/EDF5OQSH63YUYBTKOAEOXDCO2U/bundle.json","state_url":"https://pith.science/pith/EDF5OQSH63YUYBTKOAEOXDCO2U/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/EDF5OQSH63YUYBTKOAEOXDCO2U/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":"2026-06-04T04:31:23Z","links":{"resolver":"https://pith.science/pith/EDF5OQSH63YUYBTKOAEOXDCO2U","bundle":"https://pith.science/pith/EDF5OQSH63YUYBTKOAEOXDCO2U/bundle.json","state":"https://pith.science/pith/EDF5OQSH63YUYBTKOAEOXDCO2U/state.json","well_known_bundle":"https://pith.science/.well-known/pith/EDF5OQSH63YUYBTKOAEOXDCO2U/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:EDF5OQSH63YUYBTKOAEOXDCO2U","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":"68bddfce41a46d60f9c9ae6cb67140f556dee2e33f81a367a2188b9494131cde","cross_cats_sorted":["cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T18:05:44Z","title_canon_sha256":"bb8e3624248be3d14e90779a8dee6272c67d4e88957368d19422585411373f18"},"schema_version":"1.0","source":{"id":"2602.16666","kind":"arxiv","version":3}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2602.16666","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"arxiv_version","alias_value":"2602.16666v3","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2602.16666","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"pith_short_12","alias_value":"EDF5OQSH63YU","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"pith_short_16","alias_value":"EDF5OQSH63YUYBTK","created_at":"2026-06-03T01:05:10Z"},{"alias_kind":"pith_short_8","alias_value":"EDF5OQSH","created_at":"2026-06-03T01:05:10Z"}],"graph_snapshots":[{"event_id":"sha256:6fc952d54f362765a50b5b63e2169eed246552d28ef24d1b735909c47204c7ba","target":"graph","created_at":"2026-06-03T01:05:10Z","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/2602.16666/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"AI agents are increasingly deployed to execute important tasks. While rising accuracy scores on standard benchmarks suggest rapid progress, many agents still continue to fail in practice. This discrepancy highlights a fundamental limitation of current evaluations: compressing agent behavior into a single success metric obscures critical operational flaws. Notably, it ignores whether agents behave consistently across runs, withstand perturbations, fail predictably, or have bounded error severity. Grounded in safety-critical engineering, we provide a holistic performance profile by proposing twe","authors_text":"Arvind Narayanan, Kangheng Liu, Peter Kirgis, Saiteja Utpala, Sayash Kapoor, Stephan Rabanser","cross_cats":["cs.CY","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T18:05:44Z","title":"Towards a Science of AI Agent Reliability"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2602.16666","kind":"arxiv","version":3},"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":"sha256:7e1cf77e22feab8e3c47b00dadbfbd7360fe7f06c3ff14127c7aa567243bd26d","target":"record","created_at":"2026-06-03T01:05:10Z","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":"68bddfce41a46d60f9c9ae6cb67140f556dee2e33f81a367a2188b9494131cde","cross_cats_sorted":["cs.CY","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2026-02-18T18:05:44Z","title_canon_sha256":"bb8e3624248be3d14e90779a8dee6272c67d4e88957368d19422585411373f18"},"schema_version":"1.0","source":{"id":"2602.16666","kind":"arxiv","version":3}},"canonical_sha256":"20cbd74247f6f14c066a7008eb8c4ed507935363e3ad8531ecb11e67e65c56dd","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"20cbd74247f6f14c066a7008eb8c4ed507935363e3ad8531ecb11e67e65c56dd","first_computed_at":"2026-06-03T01:05:10.783069Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-03T01:05:10.783069Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"mbNtPIidtxqBzBepFF4EB8QpNFFZR/37Me4une6AIJ8Y5bY/QaYyX+aFGkdVjNnRji7YgLdRe8q+BEmDmMxwBA==","signature_status":"signed_v1","signed_at":"2026-06-03T01:05:10.783503Z","signed_message":"canonical_sha256_bytes"},"source_id":"2602.16666","source_kind":"arxiv","source_version":3}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7e1cf77e22feab8e3c47b00dadbfbd7360fe7f06c3ff14127c7aa567243bd26d","sha256:6fc952d54f362765a50b5b63e2169eed246552d28ef24d1b735909c47204c7ba"],"state_sha256":"9d32fe5d40c69b0593f18e5f600e90a16da664c260fcb8f2c835f60e3314a9d9"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"1IJjc7CIQTpWxQTuPcHzGPKoqxAlpbGnXKqxzqJwrlWmzzJp+RuAIxoV0+aambLkQSTAM6ZOLyWUI1rdY4+hDw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-04T04:31:23.967601Z","bundle_sha256":"7bdd44474bd29812d96ae41cd560876f84f596d91ff3d948ca87e993784b1c02"}}