{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:MBR7VRFMULEVTLDQLAGOYHOBN4","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"9ea90ab7b0ef06999d94cb18476a29bf9e82b6fe1adbdcf65945af50fa8ec8f5","cross_cats_sorted":["cs.MA"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-10T01:14:31Z","title_canon_sha256":"6da7331de59ccb85bda6ea8339502210a5c0d8c28baac954f714de277744def1"},"schema_version":"1.0","source":{"id":"2608.10030","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2608.10030","created_at":"2026-08-12T00:22:37Z"},{"alias_kind":"arxiv_version","alias_value":"2608.10030v1","created_at":"2026-08-12T00:22:37Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.10030","created_at":"2026-08-12T00:22:37Z"},{"alias_kind":"pith_short_12","alias_value":"MBR7VRFMULEV","created_at":"2026-08-12T00:22:37Z"},{"alias_kind":"pith_short_16","alias_value":"MBR7VRFMULEVTLDQ","created_at":"2026-08-12T00:22:37Z"},{"alias_kind":"pith_short_8","alias_value":"MBR7VRFM","created_at":"2026-08-12T00:22:37Z"}],"graph_snapshots":[{"event_id":"sha256:2fa101299b491991130492042e76a6384a3b0fb894c456e18142299a474bf790","target":"graph","created_at":"2026-08-12T00:22:37Z","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/2608.10030/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"As AI agents are increasingly deployed in complex environments, understanding their behaviors becomes critical. Yet behavioral scientific research on AI agents remains manual and labor-intensive. We introduce AEROBAT, the first multi-agent system to automate behavioral scientific research on AI agents. Given an arbitrary target behavior by its user, AEROBAT automatically executes a full pipeline of behavioral scientific research---generating hypotheses about the behavior, designing and executing controlled experiments, making behavioral assessments, analyzing the results, and writing reports. ","authors_text":"Denny Borsboom, Fanchen Bu, Hyunjin Hwang, Jaemin Yoo, Jaewan Chun, Jongha Lee, Kijung Shin, Soo Yong Lee, Taekwan Kim, Ziv Ben-Zion","cross_cats":["cs.MA"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-10T01:14:31Z","title":"Automating and Scaling Behavioral Scientific Research on AI Agents"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.10030","kind":"arxiv","version":1},"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:edf5c352a3e4c37344633d85dea91424f5ac2a9fd4251b1851a18dc14fb9f028","target":"record","created_at":"2026-08-12T00:22:37Z","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":"9ea90ab7b0ef06999d94cb18476a29bf9e82b6fe1adbdcf65945af50fa8ec8f5","cross_cats_sorted":["cs.MA"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.AI","submitted_at":"2026-08-10T01:14:31Z","title_canon_sha256":"6da7331de59ccb85bda6ea8339502210a5c0d8c28baac954f714de277744def1"},"schema_version":"1.0","source":{"id":"2608.10030","kind":"arxiv","version":1}},"canonical_sha256":"6063fac4aca2c959ac70580cec1dc16f04c443dc95f1242cc241182d1d6cddc4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6063fac4aca2c959ac70580cec1dc16f04c443dc95f1242cc241182d1d6cddc4","first_computed_at":"2026-08-12T00:22:37.870382Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-08-12T00:22:37.870382Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uRVZ2VJv5+znVQ8BspETp0UCUn6vkaeUzic5JX+PTGjHVhnGUVulCC24e0twO8oCI2VLGe0Ht55QR1EIdYVnCA==","signature_status":"signed_v1","signed_at":"2026-08-12T00:22:37.871889Z","signed_message":"canonical_sha256_bytes"},"source_id":"2608.10030","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:edf5c352a3e4c37344633d85dea91424f5ac2a9fd4251b1851a18dc14fb9f028","sha256:2fa101299b491991130492042e76a6384a3b0fb894c456e18142299a474bf790","sha256:4a505d11a84948beec71f64e0f19b83025f73c1b8cc86f3a4d15a1f048e27e1b"],"state_sha256":"a84b6010a27e9cd65d7f27d87fe55fdb1f76a799ec97fb40418377e1c7cde3c8"}