{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:6HV7V3EKJ23QU743QDRQLSTLK2","short_pith_number":"pith:6HV7V3EK","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"},"canonical_sha256":"f1ebfaec8a4eb70a7f9b80e305ca6b56b30708ed145f58413e119c6c2bf870db","source":{"kind":"arxiv","id":"2606.05174","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.05174","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"arxiv_version","alias_value":"2606.05174v1","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05174","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"pith_short_12","alias_value":"6HV7V3EKJ23Q","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"pith_short_16","alias_value":"6HV7V3EKJ23QU743","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"pith_short_8","alias_value":"6HV7V3EK","created_at":"2026-06-05T00:13:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:6HV7V3EKJ23QU743QDRQLSTLK2","target":"record","payload":{"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"},"canonical_sha256":"f1ebfaec8a4eb70a7f9b80e305ca6b56b30708ed145f58413e119c6c2bf870db","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"},"source_kind":"arxiv","source_id":"2606.05174","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-05T00:13:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6e5JFDzecteX6kZXLEEiRGMZ9RSRa678uBP7fL8P/b6k2NWo1gMuh2gymc40IbmkhN2vrB1IBIQM09U9W7E6AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T10:37:49.104945Z"},"content_sha256":"97c7bb8be5f4bbeadfba270915d34b9438d3cdcdea1b930c4c5104d43b625b5a","schema_version":"1.0","event_id":"sha256:97c7bb8be5f4bbeadfba270915d34b9438d3cdcdea1b930c4c5104d43b625b5a"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:6HV7V3EKJ23QU743QDRQLSTLK2","target":"graph","payload":{"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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-05T00:13:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"AnZNvOcGc3hLP/X23VEWhshc/3EOOI0LsMYvzvPUNwIU6fEEjzahdpOI1qjHlPiS/Ni34EE6gvV7FEkAJ/JpBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-05T10:37:49.105346Z"},"content_sha256":"49104958be1185a3317173d9a09935f2ca8a35282bb10244275f9a639f124901","schema_version":"1.0","event_id":"sha256:49104958be1185a3317173d9a09935f2ca8a35282bb10244275f9a639f124901"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2/bundle.json","state_url":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/6HV7V3EKJ23QU743QDRQLSTLK2/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-07-05T10:37:49Z","links":{"resolver":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2","bundle":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2/bundle.json","state":"https://pith.science/pith/6HV7V3EKJ23QU743QDRQLSTLK2/state.json","well_known_bundle":"https://pith.science/.well-known/pith/6HV7V3EKJ23QU743QDRQLSTLK2/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:6HV7V3EKJ23QU743QDRQLSTLK2","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":"a8b73403ecf612af3d033753237ac67b4c58609268631dbf7a23cd45a5c11d3b","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-17T03:08:55Z","title_canon_sha256":"96c16a3e0c9453e21cd814080ae58b71784b6647e48be250144490a57a1350f9"},"schema_version":"1.0","source":{"id":"2606.05174","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2606.05174","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"arxiv_version","alias_value":"2606.05174v1","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.05174","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"pith_short_12","alias_value":"6HV7V3EKJ23Q","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"pith_short_16","alias_value":"6HV7V3EKJ23QU743","created_at":"2026-06-05T00:13:47Z"},{"alias_kind":"pith_short_8","alias_value":"6HV7V3EK","created_at":"2026-06-05T00:13:47Z"}],"graph_snapshots":[{"event_id":"sha256:49104958be1185a3317173d9a09935f2ca8a35282bb10244275f9a639f124901","target":"graph","created_at":"2026-06-05T00:13:47Z","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/2606.05174/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Arash Ahmadi, Charles Nicholson, David Ebert, Mike Banad, Parisa Masnadi, Sarah Sharif","cross_cats":["cs.AI"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-17T03:08:55Z","title":"Improving Heart-Focused Medical Question Answering in LLMs via Variance-Aware Rubric Rewards with GRPO"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.05174","kind":"arxiv","version":1},"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":"sha256:97c7bb8be5f4bbeadfba270915d34b9438d3cdcdea1b930c4c5104d43b625b5a","target":"record","created_at":"2026-06-05T00:13:47Z","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":"a8b73403ecf612af3d033753237ac67b4c58609268631dbf7a23cd45a5c11d3b","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CL","submitted_at":"2026-04-17T03:08:55Z","title_canon_sha256":"96c16a3e0c9453e21cd814080ae58b71784b6647e48be250144490a57a1350f9"},"schema_version":"1.0","source":{"id":"2606.05174","kind":"arxiv","version":1}},"canonical_sha256":"f1ebfaec8a4eb70a7f9b80e305ca6b56b30708ed145f58413e119c6c2bf870db","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"f1ebfaec8a4eb70a7f9b80e305ca6b56b30708ed145f58413e119c6c2bf870db","first_computed_at":"2026-06-05T00:13:47.730302Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-05T00:13:47.730302Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"7aWy1WgE7nG1TaFAAb+iC2I3v0Jm5+YzoqaXAZQEKPdFtbRpSmeU7vR4PO30bodcKxNheUuuMaqc1JxbT/X1Bg==","signature_status":"signed_v1","signed_at":"2026-06-05T00:13:47.730733Z","signed_message":"canonical_sha256_bytes"},"source_id":"2606.05174","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:97c7bb8be5f4bbeadfba270915d34b9438d3cdcdea1b930c4c5104d43b625b5a","sha256:49104958be1185a3317173d9a09935f2ca8a35282bb10244275f9a639f124901"],"state_sha256":"40fa9348ab4b51dc51f5e56ac8338f41ddbe602cf7fc5adda79b10733efd3a01"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"KdfYHx5vXn6+H2Wjc9cxDUHECsKvYUAEQBRlc6ss761hPt12G/KprCS6w/b/dKclmPE0N+OpNcZUNqIm1bBJBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-05T10:37:49.107777Z","bundle_sha256":"658eb268a13565d6b48bf3b4d8d80f6664bf314611dcd164fb2741af10555c8a"}}