{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:LFSRL2E6BLSWVHKQYHF7I3UVD2","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":"04519072ba487fbf56d23db564ae2981651d68d44d89d1c7c6c99e8b8bb55649","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2024-11-07T22:51:47Z","title_canon_sha256":"60c1a34e462c38fdd63d1660b0189e384ace61527ddece86dc49748063311a8d"},"schema_version":"1.0","source":{"id":"2411.05231","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2411.05231","created_at":"2026-07-05T09:48:16Z"},{"alias_kind":"arxiv_version","alias_value":"2411.05231v2","created_at":"2026-07-05T09:48:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.05231","created_at":"2026-07-05T09:48:16Z"},{"alias_kind":"pith_short_12","alias_value":"LFSRL2E6BLSW","created_at":"2026-07-05T09:48:16Z"},{"alias_kind":"pith_short_16","alias_value":"LFSRL2E6BLSWVHKQ","created_at":"2026-07-05T09:48:16Z"},{"alias_kind":"pith_short_8","alias_value":"LFSRL2E6","created_at":"2026-07-05T09:48:16Z"}],"graph_snapshots":[{"event_id":"sha256:953a22f32424dd73326e11772e9ed787bc9ea0c78b1c06a699ebbe97e9e7f711","target":"graph","created_at":"2026-07-05T09:48:16Z","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/2411.05231/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Recent advances in generative artificial intelligence (AI) have shown promise in accurately grading open-ended student responses. However, few prior works have explored grading handwritten responses due to a lack of data and the challenge of combining visual and textual information. In this work, we leverage state-of-the-art multi-modal AI models, in particular GPT-4o, to automatically grade handwritten responses to college-level math exams. Using real student responses to questions in a probability theory exam, we evaluate GPT-4o's alignment with ground-truth scores from human graders using v","authors_text":"Adriana Caraeni, Alexander Scarlatos, Andrew Lan","cross_cats":["cs.CL","cs.LG"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2024-11-07T22:51:47Z","title":"Evaluating GPT-4 at Grading Handwritten Solutions in Math Exams"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.05231","kind":"arxiv","version":2},"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":"sha256:1d0054092a7151ee76b7bf68f385e3fa808a00331d66019d1bf10f89c4c9ad15","target":"record","created_at":"2026-07-05T09:48:16Z","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":"04519072ba487fbf56d23db564ae2981651d68d44d89d1c7c6c99e8b8bb55649","cross_cats_sorted":["cs.CL","cs.LG"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2024-11-07T22:51:47Z","title_canon_sha256":"60c1a34e462c38fdd63d1660b0189e384ace61527ddece86dc49748063311a8d"},"schema_version":"1.0","source":{"id":"2411.05231","kind":"arxiv","version":2}},"canonical_sha256":"596515e89e0ae56a9d50c1cbf46e951e86577e9be3ca18236633c65b430bf192","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"596515e89e0ae56a9d50c1cbf46e951e86577e9be3ca18236633c65b430bf192","first_computed_at":"2026-07-05T09:48:16.590159Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T09:48:16.590159Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"cynLrVhVB7vk4JqN1fdvgvrhgP4IdTmdlx3nF37FQdXtta6t/0rSiCgFo909WNXZVOYZOULa0nwR/hFpMUfhBw==","signature_status":"signed_v1","signed_at":"2026-07-05T09:48:16.590716Z","signed_message":"canonical_sha256_bytes"},"source_id":"2411.05231","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:1d0054092a7151ee76b7bf68f385e3fa808a00331d66019d1bf10f89c4c9ad15","sha256:953a22f32424dd73326e11772e9ed787bc9ea0c78b1c06a699ebbe97e9e7f711"],"state_sha256":"742b13e412f548fff75bb5bbaa99936ce10ea21b84e483428038968aa51e8508"}