{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:MLTIXGTUA263DOJ7T4WWRSVZ57","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":"ffbec594db9d5ff5b40a515a13ee59207520138ec69d008e5c461bfd754457c5","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-29T12:37:21Z","title_canon_sha256":"0ab98d48d5fa862a946d00bfd1f204e3a7ccaff8f73676194d2ea70e2fa450cd"},"schema_version":"1.0","source":{"id":"2305.18486","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2305.18486","created_at":"2026-07-05T06:28:18Z"},{"alias_kind":"arxiv_version","alias_value":"2305.18486v4","created_at":"2026-07-05T06:28:18Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2305.18486","created_at":"2026-07-05T06:28:18Z"},{"alias_kind":"pith_short_12","alias_value":"MLTIXGTUA263","created_at":"2026-07-05T06:28:18Z"},{"alias_kind":"pith_short_16","alias_value":"MLTIXGTUA263DOJ7","created_at":"2026-07-05T06:28:18Z"},{"alias_kind":"pith_short_8","alias_value":"MLTIXGTU","created_at":"2026-07-05T06:28:18Z"}],"graph_snapshots":[{"event_id":"sha256:447cea8e50ee3559ba0f10a57577c0c4d24caa2d612cc170bdce0550b1153bc3","target":"graph","created_at":"2026-07-05T06:28:18Z","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/2305.18486/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The development of large language models (LLMs) such as ChatGPT has brought a lot of attention recently. However, their evaluation in the benchmark academic datasets remains under-explored due to the difficulty of evaluating the generative outputs produced by this model against the ground truth. In this paper, we aim to present a thorough evaluation of ChatGPT's performance on diverse academic datasets, covering tasks like question-answering, text summarization, code generation, commonsense reasoning, mathematical problem-solving, machine translation, bias detection, and ethical considerations","authors_text":"Jimmy Xiangji Huang, Md Amran Hossen Bhuiyan, Md Tahmid Rahman Laskar, Mizanur Rahman, M Saiful Bari, Shafiq Joty","cross_cats":["cs.AI","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-29T12:37:21Z","title":"A Systematic Study and Comprehensive Evaluation of ChatGPT on Benchmark Datasets"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2305.18486","kind":"arxiv","version":4},"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":"sha256:33f911dfb4fc7afa85d963541346703eb3cf2c652733cd2c581f5a44a7c1ffc8","target":"record","created_at":"2026-07-05T06:28:18Z","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":"ffbec594db9d5ff5b40a515a13ee59207520138ec69d008e5c461bfd754457c5","cross_cats_sorted":["cs.AI","cs.LG"],"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2023-05-29T12:37:21Z","title_canon_sha256":"0ab98d48d5fa862a946d00bfd1f204e3a7ccaff8f73676194d2ea70e2fa450cd"},"schema_version":"1.0","source":{"id":"2305.18486","kind":"arxiv","version":4}},"canonical_sha256":"62e68b9a7406bdb1b93f9f2d68cab9efc65de31c4e4e2f19770698f7d23dac91","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"62e68b9a7406bdb1b93f9f2d68cab9efc65de31c4e4e2f19770698f7d23dac91","first_computed_at":"2026-07-05T06:28:18.907776Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:28:18.907776Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Ssj8chGEqIfM8uTDmqYkw1Z6hZGNnALNoX/cXqNiOMH78NZoiCmOXTKFsdbk3Pd031+OXKPWGFkCqqOVt/gtBQ==","signature_status":"signed_v1","signed_at":"2026-07-05T06:28:18.908216Z","signed_message":"canonical_sha256_bytes"},"source_id":"2305.18486","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:33f911dfb4fc7afa85d963541346703eb3cf2c652733cd2c581f5a44a7c1ffc8","sha256:447cea8e50ee3559ba0f10a57577c0c4d24caa2d612cc170bdce0550b1153bc3"],"state_sha256":"1fa051d14d0e8ce2878a9359c155970f5b8cea0d06afd175d271d578ff97020e"}