{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:6HV7V3EKJ23QU743QDRQLSTLK2","short_pith_number":"pith:6HV7V3EK","schema_version":"1.0","canonical_sha256":"f1ebfaec8a4eb70a7f9b80e305ca6b56b30708ed145f58413e119c6c2bf870db","source":{"kind":"arxiv","id":"2606.05174","version":1},"attestation_state":"computed","paper":{"title":"Improving Heart-Focused Medical Question Answering in LLMs via Variance-Aware Rubric Rewards with GRPO","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Arash Ahmadi, Charles Nicholson, David Ebert, Mike Banad, Parisa Masnadi, Sarah Sharif","submitted_at":"2026-04-17T03:08:55Z","abstract_excerpt":"Large Language Models (LLMs) have shown strong promise in healthcare applications. Yet deploying general-purpose models in real-world settings remains difficult due to data privacy constraints, inference costs, and limited suitability for edge or on-device use. These challenges motivate the development of smaller, more efficient models that require robust post-training strategies to ensure reliable medical reasoning. In this work, we investigate Group Relative Policy Optimization (GRPO) for post-training LLMs on heart-focused medical question answering with rubric-based supervision derived fro"},"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":"2606.05174","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-17T03:08:55Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"96c16a3e0c9453e21cd814080ae58b71784b6647e48be250144490a57a1350f9","abstract_canon_sha256":"a8b73403ecf612af3d033753237ac67b4c58609268631dbf7a23cd45a5c11d3b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-05T00:13:47.730733Z","signature_b64":"7aWy1WgE7nG1TaFAAb+iC2I3v0Jm5+YzoqaXAZQEKPdFtbRpSmeU7vR4PO30bodcKxNheUuuMaqc1JxbT/X1Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f1ebfaec8a4eb70a7f9b80e305ca6b56b30708ed145f58413e119c6c2bf870db","last_reissued_at":"2026-06-05T00:13:47.730302Z","signature_status":"signed_v1","first_computed_at":"2026-06-05T00:13:47.730302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improving Heart-Focused Medical Question Answering in LLMs via Variance-Aware Rubric Rewards with GRPO","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Arash Ahmadi, Charles Nicholson, David Ebert, Mike Banad, Parisa Masnadi, Sarah Sharif","submitted_at":"2026-04-17T03:08:55Z","abstract_excerpt":"Large Language Models (LLMs) have shown strong promise in healthcare applications. Yet deploying general-purpose models in real-world settings remains difficult due to data privacy constraints, inference costs, and limited suitability for edge or on-device use. These challenges motivate the development of smaller, more efficient models that require robust post-training strategies to ensure reliable medical reasoning. In this work, we investigate Group Relative Policy Optimization (GRPO) for post-training LLMs on heart-focused medical question answering with rubric-based supervision derived fro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05174","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/2606.05174/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":"2606.05174","created_at":"2026-06-05T00:13:47.730372+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.05174v1","created_at":"2026-06-05T00:13:47.730372+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05174","created_at":"2026-06-05T00:13:47.730372+00:00"},{"alias_kind":"pith_short_12","alias_value":"6HV7V3EKJ23Q","created_at":"2026-06-05T00:13:47.730372+00:00"},{"alias_kind":"pith_short_16","alias_value":"6HV7V3EKJ23QU743","created_at":"2026-06-05T00:13:47.730372+00:00"},{"alias_kind":"pith_short_8","alias_value":"6HV7V3EK","created_at":"2026-06-05T00:13:47.730372+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2","json":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2.json","graph_json":"https://pith.science/api/pith-number/6HV7V3EKJ23QU743QDRQLSTLK2/graph.json","events_json":"https://pith.science/api/pith-number/6HV7V3EKJ23QU743QDRQLSTLK2/events.json","paper":"https://pith.science/paper/6HV7V3EK"},"agent_actions":{"view_html":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2","download_json":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2.json","view_paper":"https://pith.science/paper/6HV7V3EK","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.05174&json=true","fetch_graph":"https://pith.science/api/pith-number/6HV7V3EKJ23QU743QDRQLSTLK2/graph.json","fetch_events":"https://pith.science/api/pith-number/6HV7V3EKJ23QU743QDRQLSTLK2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2/action/storage_attestation","attest_author":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2/action/author_attestation","sign_citation":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2/action/citation_signature","submit_replication":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2/action/replication_record"}},"created_at":"2026-06-05T00:13:47.730372+00:00","updated_at":"2026-06-05T00:13:47.730372+00:00"}