{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:P2DC7CTHVXI4CCPLU5UKTYMVZX","short_pith_number":"pith:P2DC7CTH","schema_version":"1.0","canonical_sha256":"7e862f8a67add1c109eba768a9e195cde1fe9be2ce41f596dad3cf3f0ab5d4b8","source":{"kind":"arxiv","id":"2312.11524","version":1},"attestation_state":"computed","paper":{"title":"Assessing GPT4-V on Structured Reasoning Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Gust Verbruggen, Jos\\'e Cambronero, Mukul Singh, Sumit Gulwani, Vu Le","submitted_at":"2023-12-13T08:54:49Z","abstract_excerpt":"Multi-modality promises to unlock further uses for large language models. Recently, the state-of-the-art language model GPT-4 was enhanced with vision capabilities. We carry out a prompting evaluation of GPT-4V and five other baselines on structured reasoning tasks, such as mathematical reasoning, visual data analysis, and code generation. We show that visual Chain-of-Thought, an extension of Chain-of-Thought to multi-modal LLMs, yields significant improvements over the vanilla model. We also present a categorized analysis of scenarios where these models perform well and where they struggle, h"},"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":"2312.11524","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-12-13T08:54:49Z","cross_cats_sorted":["cs.AI","cs.CV"],"title_canon_sha256":"10fe26f5d464289c03790859042e80a0a3d6ce7127fe29be833c7c09e71d8bd4","abstract_canon_sha256":"77fab89cf75859b726a15ece8ba3796c86f2e6b3cbb6fe2fbe3f43b8309bc1c3"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:25:50.819898Z","signature_b64":"3HSRCMOZ1K7Fy0Vi+ZjFolt/+fJAcGtdLEspH12f7zHpKZUJSs65yD0LOcknIcoot9POg+XI2HZaJLseDhvMDw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"7e862f8a67add1c109eba768a9e195cde1fe9be2ce41f596dad3cf3f0ab5d4b8","last_reissued_at":"2026-07-05T07:25:50.819448Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:25:50.819448Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Assessing GPT4-V on Structured Reasoning Tasks","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.CV"],"primary_cat":"cs.CL","authors_text":"Gust Verbruggen, Jos\\'e Cambronero, Mukul Singh, Sumit Gulwani, Vu Le","submitted_at":"2023-12-13T08:54:49Z","abstract_excerpt":"Multi-modality promises to unlock further uses for large language models. Recently, the state-of-the-art language model GPT-4 was enhanced with vision capabilities. We carry out a prompting evaluation of GPT-4V and five other baselines on structured reasoning tasks, such as mathematical reasoning, visual data analysis, and code generation. We show that visual Chain-of-Thought, an extension of Chain-of-Thought to multi-modal LLMs, yields significant improvements over the vanilla model. We also present a categorized analysis of scenarios where these models perform well and where they struggle, h"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.11524","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/2312.11524/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":"2312.11524","created_at":"2026-07-05T07:25:50.819509+00:00"},{"alias_kind":"arxiv_version","alias_value":"2312.11524v1","created_at":"2026-07-05T07:25:50.819509+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.11524","created_at":"2026-07-05T07:25:50.819509+00:00"},{"alias_kind":"pith_short_12","alias_value":"P2DC7CTHVXI4","created_at":"2026-07-05T07:25:50.819509+00:00"},{"alias_kind":"pith_short_16","alias_value":"P2DC7CTHVXI4CCPL","created_at":"2026-07-05T07:25:50.819509+00:00"},{"alias_kind":"pith_short_8","alias_value":"P2DC7CTH","created_at":"2026-07-05T07:25:50.819509+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2511.23253","citing_title":"AgroCoT: A Chain-of-Thought Benchmark for Evaluating Reasoning in Vision-Language Models for Agriculture","ref_index":39,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/P2DC7CTHVXI4CCPLU5UKTYMVZX","json":"https://pith.science/pith/P2DC7CTHVXI4CCPLU5UKTYMVZX.json","graph_json":"https://pith.science/api/pith-number/P2DC7CTHVXI4CCPLU5UKTYMVZX/graph.json","events_json":"https://pith.science/api/pith-number/P2DC7CTHVXI4CCPLU5UKTYMVZX/events.json","paper":"https://pith.science/paper/P2DC7CTH"},"agent_actions":{"view_html":"https://pith.science/pith/P2DC7CTHVXI4CCPLU5UKTYMVZX","download_json":"https://pith.science/pith/P2DC7CTHVXI4CCPLU5UKTYMVZX.json","view_paper":"https://pith.science/paper/P2DC7CTH","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2312.11524&json=true","fetch_graph":"https://pith.science/api/pith-number/P2DC7CTHVXI4CCPLU5UKTYMVZX/graph.json","fetch_events":"https://pith.science/api/pith-number/P2DC7CTHVXI4CCPLU5UKTYMVZX/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/P2DC7CTHVXI4CCPLU5UKTYMVZX/action/timestamp_anchor","attest_storage":"https://pith.science/pith/P2DC7CTHVXI4CCPLU5UKTYMVZX/action/storage_attestation","attest_author":"https://pith.science/pith/P2DC7CTHVXI4CCPLU5UKTYMVZX/action/author_attestation","sign_citation":"https://pith.science/pith/P2DC7CTHVXI4CCPLU5UKTYMVZX/action/citation_signature","submit_replication":"https://pith.science/pith/P2DC7CTHVXI4CCPLU5UKTYMVZX/action/replication_record"}},"created_at":"2026-07-05T07:25:50.819509+00:00","updated_at":"2026-07-05T07:25:50.819509+00:00"}