{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:H7AMOGA3LLD3YO73PYOGQ5RJG3","short_pith_number":"pith:H7AMOGA3","canonical_record":{"source":{"id":"2304.05372","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-09T04:53:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"626dccefc240b144a46a74c39c5766fe4e9d2898267667a2dc65425492c714a3","abstract_canon_sha256":"f4ce70b4e5361219077a21535584529aa104555bb599733855e7fdd7c45cf87f"},"schema_version":"1.0"},"canonical_sha256":"3fc0c7181b5ac7bc3bfb7e1c68762936e2d6f12e8d07c4e3d8b9456fa87c30ce","source":{"kind":"arxiv","id":"2304.05372","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2304.05372","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"arxiv_version","alias_value":"2304.05372v1","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.05372","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"pith_short_12","alias_value":"H7AMOGA3LLD3","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"pith_short_16","alias_value":"H7AMOGA3LLD3YO73","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"pith_short_8","alias_value":"H7AMOGA3","created_at":"2026-07-05T06:08:55Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:H7AMOGA3LLD3YO73PYOGQ5RJG3","target":"record","payload":{"canonical_record":{"source":{"id":"2304.05372","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-09T04:53:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"626dccefc240b144a46a74c39c5766fe4e9d2898267667a2dc65425492c714a3","abstract_canon_sha256":"f4ce70b4e5361219077a21535584529aa104555bb599733855e7fdd7c45cf87f"},"schema_version":"1.0"},"canonical_sha256":"3fc0c7181b5ac7bc3bfb7e1c68762936e2d6f12e8d07c4e3d8b9456fa87c30ce","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:08:55.230426Z","signature_b64":"fQKZ1a/pbehQAvNqgi9+UXlQ6e1U8t03s7HbCUBOknSNLBfoo4Umymw8ESUkOcOjA0uJHb8jcLi2kcTzE3NqDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3fc0c7181b5ac7bc3bfb7e1c68762936e2d6f12e8d07c4e3d8b9456fa87c30ce","last_reissued_at":"2026-07-05T06:08:55.230033Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:08:55.230033Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2304.05372","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T06:08:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"RnWnDs/xdvL9bb4iMiM9bd1H6TWSuwgWrjywqtxcV86AreTURdvt8OX+yH6vy7MxfOWWUlTp4pmEAbM2b4DlDQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-29T02:22:09.164628Z"},"content_sha256":"10b06b6379fe9dc53cd7f2d47b375bfe00b0afba97db640904e52d2d3144e6eb","schema_version":"1.0","event_id":"sha256:10b06b6379fe9dc53cd7f2d47b375bfe00b0afba97db640904e52d2d3144e6eb"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:H7AMOGA3LLD3YO73PYOGQ5RJG3","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Can ChatGPT and Bard Generate Aligned Assessment Items? A Reliability Analysis against Human Performance","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Abdolvahab Khademi","submitted_at":"2023-04-09T04:53:15Z","abstract_excerpt":"ChatGPT and Bard are AI chatbots based on Large Language Models (LLM) that are slated to promise different applications in diverse areas. In education, these AI technologies have been tested for applications in assessment and teaching. In assessment, AI has long been used in automated essay scoring and automated item generation. One psychometric property that these tools must have to assist or replace humans in assessment is high reliability in terms of agreement between AI scores and human raters. In this paper, we measure the reliability of OpenAI ChatGP and Google Bard LLMs tools against ex"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.05372","kind":"arxiv","version":1},"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/2304.05372/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":"2026-07-05T06:08:55Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Yfec3pFmnV69/kATwi4onGmlMHdVrM+lNTHAavocCYAkp+AY+EB4ORtBghHdbRj1PZv4br+Cn+sFEbEFjQlxBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-29T02:22:09.165026Z"},"content_sha256":"0689141c74345f9d51722f3693f210a00666e6bb98c2516447257990290d4c32","schema_version":"1.0","event_id":"sha256:0689141c74345f9d51722f3693f210a00666e6bb98c2516447257990290d4c32"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/H7AMOGA3LLD3YO73PYOGQ5RJG3/bundle.json","state_url":"https://pith.science/pith/H7AMOGA3LLD3YO73PYOGQ5RJG3/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/H7AMOGA3LLD3YO73PYOGQ5RJG3/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":"2026-07-29T02:22:09Z","links":{"resolver":"https://pith.science/pith/H7AMOGA3LLD3YO73PYOGQ5RJG3","bundle":"https://pith.science/pith/H7AMOGA3LLD3YO73PYOGQ5RJG3/bundle.json","state":"https://pith.science/pith/H7AMOGA3LLD3YO73PYOGQ5RJG3/state.json","well_known_bundle":"https://pith.science/.well-known/pith/H7AMOGA3LLD3YO73PYOGQ5RJG3/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:H7AMOGA3LLD3YO73PYOGQ5RJG3","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":"f4ce70b4e5361219077a21535584529aa104555bb599733855e7fdd7c45cf87f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-09T04:53:15Z","title_canon_sha256":"626dccefc240b144a46a74c39c5766fe4e9d2898267667a2dc65425492c714a3"},"schema_version":"1.0","source":{"id":"2304.05372","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2304.05372","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"arxiv_version","alias_value":"2304.05372v1","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.05372","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"pith_short_12","alias_value":"H7AMOGA3LLD3","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"pith_short_16","alias_value":"H7AMOGA3LLD3YO73","created_at":"2026-07-05T06:08:55Z"},{"alias_kind":"pith_short_8","alias_value":"H7AMOGA3","created_at":"2026-07-05T06:08:55Z"}],"graph_snapshots":[{"event_id":"sha256:0689141c74345f9d51722f3693f210a00666e6bb98c2516447257990290d4c32","target":"graph","created_at":"2026-07-05T06:08:55Z","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/2304.05372/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"ChatGPT and Bard are AI chatbots based on Large Language Models (LLM) that are slated to promise different applications in diverse areas. In education, these AI technologies have been tested for applications in assessment and teaching. In assessment, AI has long been used in automated essay scoring and automated item generation. One psychometric property that these tools must have to assist or replace humans in assessment is high reliability in terms of agreement between AI scores and human raters. In this paper, we measure the reliability of OpenAI ChatGP and Google Bard LLMs tools against ex","authors_text":"Abdolvahab Khademi","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-09T04:53:15Z","title":"Can ChatGPT and Bard Generate Aligned Assessment Items? A Reliability Analysis against Human Performance"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.05372","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":"sha256:10b06b6379fe9dc53cd7f2d47b375bfe00b0afba97db640904e52d2d3144e6eb","target":"record","created_at":"2026-07-05T06:08:55Z","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":"f4ce70b4e5361219077a21535584529aa104555bb599733855e7fdd7c45cf87f","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by-nc-nd/4.0/","primary_cat":"cs.CL","submitted_at":"2023-04-09T04:53:15Z","title_canon_sha256":"626dccefc240b144a46a74c39c5766fe4e9d2898267667a2dc65425492c714a3"},"schema_version":"1.0","source":{"id":"2304.05372","kind":"arxiv","version":1}},"canonical_sha256":"3fc0c7181b5ac7bc3bfb7e1c68762936e2d6f12e8d07c4e3d8b9456fa87c30ce","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3fc0c7181b5ac7bc3bfb7e1c68762936e2d6f12e8d07c4e3d8b9456fa87c30ce","first_computed_at":"2026-07-05T06:08:55.230033Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T06:08:55.230033Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fQKZ1a/pbehQAvNqgi9+UXlQ6e1U8t03s7HbCUBOknSNLBfoo4Umymw8ESUkOcOjA0uJHb8jcLi2kcTzE3NqDg==","signature_status":"signed_v1","signed_at":"2026-07-05T06:08:55.230426Z","signed_message":"canonical_sha256_bytes"},"source_id":"2304.05372","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:10b06b6379fe9dc53cd7f2d47b375bfe00b0afba97db640904e52d2d3144e6eb","sha256:0689141c74345f9d51722f3693f210a00666e6bb98c2516447257990290d4c32"],"state_sha256":"55137b80095c37d0e22fcbad2d4aafcde48449258308e3040da6f4c9f0e979a3"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"lBrYd4A3UN7iSW3n9L/MCRWH5COmOxJ81dug7hdxA+BEj6CTMW8pGkMrqDs4EXRi0n5+gBVCYqiB1IKi5yypBQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-29T02:22:09.167285Z","bundle_sha256":"c64bedc4bd42814019518e8e533cc7db4e071bfde738b4b5fd3b29dc3ab79807"}}