{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:TONP7GQOK2OJWN7NEK36XNOFOZ","short_pith_number":"pith:TONP7GQO","schema_version":"1.0","canonical_sha256":"9b9aff9a0e569c9b37ed22b7ebb5c576648582a5af5471d52df7230e68583b80","source":{"kind":"arxiv","id":"2311.13735","version":1},"attestation_state":"computed","paper":{"title":"Surpassing GPT-4 Medical Coding with a Two-Stage Approach","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Eran Halperin, Joel Stremmel, Sanjit Singh Batra, Zhichao Yang","submitted_at":"2023-11-22T23:35:13Z","abstract_excerpt":"Recent advances in large language models (LLMs) show potential for clinical applications, such as clinical decision support and trial recommendations. However, the GPT-4 LLM predicts an excessive number of ICD codes for medical coding tasks, leading to high recall but low precision. To tackle this challenge, we introduce LLM-codex, a two-stage approach to predict ICD codes that first generates evidence proposals using an LLM and then employs an LSTM-based verification stage. The LSTM learns from both the LLM's high recall and human expert's high precision, using a custom loss function. Our mod"},"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":"2311.13735","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2023-11-22T23:35:13Z","cross_cats_sorted":[],"title_canon_sha256":"13527e06d7df7f7c6c4a843b4d9a0fbf25c2013a6a605a59c07878a6080bd3c8","abstract_canon_sha256":"65a8b6e1dee7deeea234131669b5d118f35fc4217e78349621f354bf4e9ce11d"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:16:07.086928Z","signature_b64":"Mk4LKvjZWpfkyeCymLA69EYroux9zTRLqR9H0ST+hV22/vdo9ISiQorqerRGH3UE6of961mS119ku7FI5IhHCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"9b9aff9a0e569c9b37ed22b7ebb5c576648582a5af5471d52df7230e68583b80","last_reissued_at":"2026-07-05T07:16:07.086430Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:16:07.086430Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Surpassing GPT-4 Medical Coding with a Two-Stage Approach","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Eran Halperin, Joel Stremmel, Sanjit Singh Batra, Zhichao Yang","submitted_at":"2023-11-22T23:35:13Z","abstract_excerpt":"Recent advances in large language models (LLMs) show potential for clinical applications, such as clinical decision support and trial recommendations. However, the GPT-4 LLM predicts an excessive number of ICD codes for medical coding tasks, leading to high recall but low precision. To tackle this challenge, we introduce LLM-codex, a two-stage approach to predict ICD codes that first generates evidence proposals using an LLM and then employs an LSTM-based verification stage. The LSTM learns from both the LLM's high recall and human expert's high precision, using a custom loss function. Our mod"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2311.13735","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/2311.13735/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":"2311.13735","created_at":"2026-07-05T07:16:07.086495+00:00"},{"alias_kind":"arxiv_version","alias_value":"2311.13735v1","created_at":"2026-07-05T07:16:07.086495+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2311.13735","created_at":"2026-07-05T07:16:07.086495+00:00"},{"alias_kind":"pith_short_12","alias_value":"TONP7GQOK2OJ","created_at":"2026-07-05T07:16:07.086495+00:00"},{"alias_kind":"pith_short_16","alias_value":"TONP7GQOK2OJWN7N","created_at":"2026-07-05T07:16:07.086495+00:00"},{"alias_kind":"pith_short_8","alias_value":"TONP7GQO","created_at":"2026-07-05T07:16:07.086495+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.21154","citing_title":"Automated ICD Classification of Psychiatric Diagnoses: From Classical NLP to Large Language Models","ref_index":25,"is_internal_anchor":false},{"citing_arxiv_id":"2605.17755","citing_title":"Bridging the Version Gap: Multi-version Training Improves ICD Code Prediction, Especially for Rare Codes","ref_index":7,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/TONP7GQOK2OJWN7NEK36XNOFOZ","json":"https://pith.science/pith/TONP7GQOK2OJWN7NEK36XNOFOZ.json","graph_json":"https://pith.science/api/pith-number/TONP7GQOK2OJWN7NEK36XNOFOZ/graph.json","events_json":"https://pith.science/api/pith-number/TONP7GQOK2OJWN7NEK36XNOFOZ/events.json","paper":"https://pith.science/paper/TONP7GQO"},"agent_actions":{"view_html":"https://pith.science/pith/TONP7GQOK2OJWN7NEK36XNOFOZ","download_json":"https://pith.science/pith/TONP7GQOK2OJWN7NEK36XNOFOZ.json","view_paper":"https://pith.science/paper/TONP7GQO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2311.13735&json=true","fetch_graph":"https://pith.science/api/pith-number/TONP7GQOK2OJWN7NEK36XNOFOZ/graph.json","fetch_events":"https://pith.science/api/pith-number/TONP7GQOK2OJWN7NEK36XNOFOZ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/TONP7GQOK2OJWN7NEK36XNOFOZ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/TONP7GQOK2OJWN7NEK36XNOFOZ/action/storage_attestation","attest_author":"https://pith.science/pith/TONP7GQOK2OJWN7NEK36XNOFOZ/action/author_attestation","sign_citation":"https://pith.science/pith/TONP7GQOK2OJWN7NEK36XNOFOZ/action/citation_signature","submit_replication":"https://pith.science/pith/TONP7GQOK2OJWN7NEK36XNOFOZ/action/replication_record"}},"created_at":"2026-07-05T07:16:07.086495+00:00","updated_at":"2026-07-05T07:16:07.086495+00:00"}