{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:GEQACNLEPUKNOFWXP4OHTVKV6D","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":"9131e596a96dfc0a32be381195391867cab37768d5fe65e86104bf3a27bf77d4","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-04T01:07:20Z","title_canon_sha256":"fe592830912fdc8e2a57423c9208a66ffae6eaa7b76310585ec6f0f712606bee"},"schema_version":"1.0","source":{"id":"2306.02224","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2306.02224","created_at":"2026-07-05T06:17:14Z"},{"alias_kind":"arxiv_version","alias_value":"2306.02224v1","created_at":"2026-07-05T06:17:14Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.02224","created_at":"2026-07-05T06:17:14Z"},{"alias_kind":"pith_short_12","alias_value":"GEQACNLEPUKN","created_at":"2026-07-05T06:17:14Z"},{"alias_kind":"pith_short_16","alias_value":"GEQACNLEPUKNOFWX","created_at":"2026-07-05T06:17:14Z"},{"alias_kind":"pith_short_8","alias_value":"GEQACNLE","created_at":"2026-07-05T06:17:14Z"}],"graph_snapshots":[{"event_id":"sha256:21279ae91179f136ed7e54e9b1b455d9fd8eaf2b97aa319c3d063fc2c4730562","target":"graph","created_at":"2026-07-05T06:17:14Z","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/2306.02224/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Auto-GPT is an autonomous agent that leverages recent advancements in adapting Large Language Models (LLMs) for decision-making tasks. While there has been a growing interest in Auto-GPT stypled agents, questions remain regarding the effectiveness and flexibility of Auto-GPT in solving real-world decision-making tasks. Its limited capability for real-world engagement and the absence of benchmarks contribute to these uncertainties. In this paper, we present a comprehensive benchmark study of Auto-GPT styled agents in decision-making tasks that simulate real-world scenarios. Our aim is to gain d","authors_text":"Hui Yang, Sifu Yue, Yunzhong He","cross_cats":["cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-04T01:07:20Z","title":"Auto-GPT for Online Decision Making: Benchmarks and Additional Opinions"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.02224","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":"sha256:692f570a619ec7be5360912aab42efbbe7daac6081bbc38124c8366948617ef2","target":"record","created_at":"2026-07-05T06:17:14Z","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":"9131e596a96dfc0a32be381195391867cab37768d5fe65e86104bf3a27bf77d4","cross_cats_sorted":["cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.AI","submitted_at":"2023-06-04T01:07:20Z","title_canon_sha256":"fe592830912fdc8e2a57423c9208a66ffae6eaa7b76310585ec6f0f712606bee"},"schema_version":"1.0","source":{"id":"2306.02224","kind":"arxiv","version":1}},"canonical_sha256":"31200135647d14d716d77f1c79d555f0f86d271fb0de62ca45c335e998a24af2","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"31200135647d14d716d77f1c79d555f0f86d271fb0de62ca45c335e998a24af2","first_computed_at":"2026-07-05T06:17:14.461041Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:17:14.461041Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"bLIGQsLIMgRYzVX/1LY2NtUHyM9ZJZdSf0XUbSvywdP5JhdUWb47FoRB2zDMDLBHplkMsJ2ZVZRq4PpYT3zqCA==","signature_status":"signed_v1","signed_at":"2026-07-05T06:17:14.461448Z","signed_message":"canonical_sha256_bytes"},"source_id":"2306.02224","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:692f570a619ec7be5360912aab42efbbe7daac6081bbc38124c8366948617ef2","sha256:21279ae91179f136ed7e54e9b1b455d9fd8eaf2b97aa319c3d063fc2c4730562"],"state_sha256":"e9b740eaebdae8627865d6423d4326ea4481ede4c5cfd19bd610bc044c6b008f"}