{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:LFSRL2E6BLSWVHKQYHF7I3UVD2","short_pith_number":"pith:LFSRL2E6","schema_version":"1.0","canonical_sha256":"596515e89e0ae56a9d50c1cbf46e951e86577e9be3ca18236633c65b430bf192","source":{"kind":"arxiv","id":"2411.05231","version":2},"attestation_state":"computed","paper":{"title":"Evaluating GPT-4 at Grading Handwritten Solutions in Math Exams","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CY","authors_text":"Adriana Caraeni, Alexander Scarlatos, Andrew Lan","submitted_at":"2024-11-07T22:51:47Z","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"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2411.05231","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CY","submitted_at":"2024-11-07T22:51:47Z","cross_cats_sorted":["cs.CL","cs.LG"],"title_canon_sha256":"60c1a34e462c38fdd63d1660b0189e384ace61527ddece86dc49748063311a8d","abstract_canon_sha256":"04519072ba487fbf56d23db564ae2981651d68d44d89d1c7c6c99e8b8bb55649"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T09:48:16.590716Z","signature_b64":"cynLrVhVB7vk4JqN1fdvgvrhgP4IdTmdlx3nF37FQdXtta6t/0rSiCgFo909WNXZVOYZOULa0nwR/hFpMUfhBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"596515e89e0ae56a9d50c1cbf46e951e86577e9be3ca18236633c65b430bf192","last_reissued_at":"2026-07-05T09:48:16.590159Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T09:48:16.590159Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Evaluating GPT-4 at Grading Handwritten Solutions in Math Exams","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CL","cs.LG"],"primary_cat":"cs.CY","authors_text":"Adriana Caraeni, Alexander Scarlatos, Andrew Lan","submitted_at":"2024-11-07T22:51:47Z","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"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2411.05231","kind":"arxiv","version":2},"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/2411.05231/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2411.05231","created_at":"2026-07-05T09:48:16.590225+00:00"},{"alias_kind":"arxiv_version","alias_value":"2411.05231v2","created_at":"2026-07-05T09:48:16.590225+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2411.05231","created_at":"2026-07-05T09:48:16.590225+00:00"},{"alias_kind":"pith_short_12","alias_value":"LFSRL2E6BLSW","created_at":"2026-07-05T09:48:16.590225+00:00"},{"alias_kind":"pith_short_16","alias_value":"LFSRL2E6BLSWVHKQ","created_at":"2026-07-05T09:48:16.590225+00:00"},{"alias_kind":"pith_short_8","alias_value":"LFSRL2E6","created_at":"2026-07-05T09:48:16.590225+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.24973","citing_title":"LLM Performance on a Real, Double-Marked GCSE Benchmark","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LFSRL2E6BLSWVHKQYHF7I3UVD2","json":"https://pith.science/pith/LFSRL2E6BLSWVHKQYHF7I3UVD2.json","graph_json":"https://pith.science/api/pith-number/LFSRL2E6BLSWVHKQYHF7I3UVD2/graph.json","events_json":"https://pith.science/api/pith-number/LFSRL2E6BLSWVHKQYHF7I3UVD2/events.json","paper":"https://pith.science/paper/LFSRL2E6"},"agent_actions":{"view_html":"https://pith.science/pith/LFSRL2E6BLSWVHKQYHF7I3UVD2","download_json":"https://pith.science/pith/LFSRL2E6BLSWVHKQYHF7I3UVD2.json","view_paper":"https://pith.science/paper/LFSRL2E6","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2411.05231&json=true","fetch_graph":"https://pith.science/api/pith-number/LFSRL2E6BLSWVHKQYHF7I3UVD2/graph.json","fetch_events":"https://pith.science/api/pith-number/LFSRL2E6BLSWVHKQYHF7I3UVD2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LFSRL2E6BLSWVHKQYHF7I3UVD2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LFSRL2E6BLSWVHKQYHF7I3UVD2/action/storage_attestation","attest_author":"https://pith.science/pith/LFSRL2E6BLSWVHKQYHF7I3UVD2/action/author_attestation","sign_citation":"https://pith.science/pith/LFSRL2E6BLSWVHKQYHF7I3UVD2/action/citation_signature","submit_replication":"https://pith.science/pith/LFSRL2E6BLSWVHKQYHF7I3UVD2/action/replication_record"}},"created_at":"2026-07-05T09:48:16.590225+00:00","updated_at":"2026-07-05T09:48:16.590225+00:00"}