{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:A2CWY3QMLJUTNEYMHLDPP6JC34","short_pith_number":"pith:A2CWY3QM","schema_version":"1.0","canonical_sha256":"06856c6e0c5a6936930c3ac6f7f922df30e1ff9926b407573830991b6dc78ed1","source":{"kind":"arxiv","id":"2506.06600","version":2},"attestation_state":"computed","paper":{"title":"RARL: Improving Medical VLM Reasoning and Generalization with Reinforcement Learning and LoRA under Data and Hardware Constraints","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chris Ngo, Tan-Hanh Pham","submitted_at":"2025-06-07T00:26:23Z","abstract_excerpt":"The growing integration of vision-language models (VLMs) in medical applications offers promising support for diagnostic reasoning. However, current medical VLMs often face limitations in generalization, transparency, and computational efficiency-barriers that hinder deployment in real-world, resource-constrained settings. To address these challenges, we propose a Reasoning-Aware Reinforcement Learning framework, \\textbf{RARL}, that enhances the reasoning capabilities of medical VLMs while remaining efficient and adaptable to low-resource environments. Our approach fine-tunes a lightweight bas"},"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":"2506.06600","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CV","submitted_at":"2025-06-07T00:26:23Z","cross_cats_sorted":[],"title_canon_sha256":"fcbd06e4379ca83ea856dcbfdf4f5fd9b32ae53957f682c9034345a94938ddcc","abstract_canon_sha256":"4c72e0a049a00fe7a1972c6406081b0cd0467fbac7a64cf7a57eee1cffb4dbab"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:21:48.368207Z","signature_b64":"H9RnhSX3lzmadm8puVZFz3u1Gl0KybNhsPDdgnHmsbQjji7I9wSEqBwk6nQHxMa9oLmJtfsK8s4pMTSKT1HlAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"06856c6e0c5a6936930c3ac6f7f922df30e1ff9926b407573830991b6dc78ed1","last_reissued_at":"2026-07-05T11:21:48.367721Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:21:48.367721Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"RARL: Improving Medical VLM Reasoning and Generalization with Reinforcement Learning and LoRA under Data and Hardware Constraints","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Chris Ngo, Tan-Hanh Pham","submitted_at":"2025-06-07T00:26:23Z","abstract_excerpt":"The growing integration of vision-language models (VLMs) in medical applications offers promising support for diagnostic reasoning. However, current medical VLMs often face limitations in generalization, transparency, and computational efficiency-barriers that hinder deployment in real-world, resource-constrained settings. To address these challenges, we propose a Reasoning-Aware Reinforcement Learning framework, \\textbf{RARL}, that enhances the reasoning capabilities of medical VLMs while remaining efficient and adaptable to low-resource environments. Our approach fine-tunes a lightweight bas"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.06600","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/2506.06600/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":"2506.06600","created_at":"2026-07-05T11:21:48.367781+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.06600v2","created_at":"2026-07-05T11:21:48.367781+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.06600","created_at":"2026-07-05T11:21:48.367781+00:00"},{"alias_kind":"pith_short_12","alias_value":"A2CWY3QMLJUT","created_at":"2026-07-05T11:21:48.367781+00:00"},{"alias_kind":"pith_short_16","alias_value":"A2CWY3QMLJUTNEYM","created_at":"2026-07-05T11:21:48.367781+00:00"},{"alias_kind":"pith_short_8","alias_value":"A2CWY3QM","created_at":"2026-07-05T11:21:48.367781+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.25273","citing_title":"LLM-as-a-Judge in Healthcare: A Scoping Analysis of Applications, Methods, and Human Alignment","ref_index":116,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/A2CWY3QMLJUTNEYMHLDPP6JC34","json":"https://pith.science/pith/A2CWY3QMLJUTNEYMHLDPP6JC34.json","graph_json":"https://pith.science/api/pith-number/A2CWY3QMLJUTNEYMHLDPP6JC34/graph.json","events_json":"https://pith.science/api/pith-number/A2CWY3QMLJUTNEYMHLDPP6JC34/events.json","paper":"https://pith.science/paper/A2CWY3QM"},"agent_actions":{"view_html":"https://pith.science/pith/A2CWY3QMLJUTNEYMHLDPP6JC34","download_json":"https://pith.science/pith/A2CWY3QMLJUTNEYMHLDPP6JC34.json","view_paper":"https://pith.science/paper/A2CWY3QM","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.06600&json=true","fetch_graph":"https://pith.science/api/pith-number/A2CWY3QMLJUTNEYMHLDPP6JC34/graph.json","fetch_events":"https://pith.science/api/pith-number/A2CWY3QMLJUTNEYMHLDPP6JC34/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/A2CWY3QMLJUTNEYMHLDPP6JC34/action/timestamp_anchor","attest_storage":"https://pith.science/pith/A2CWY3QMLJUTNEYMHLDPP6JC34/action/storage_attestation","attest_author":"https://pith.science/pith/A2CWY3QMLJUTNEYMHLDPP6JC34/action/author_attestation","sign_citation":"https://pith.science/pith/A2CWY3QMLJUTNEYMHLDPP6JC34/action/citation_signature","submit_replication":"https://pith.science/pith/A2CWY3QMLJUTNEYMHLDPP6JC34/action/replication_record"}},"created_at":"2026-07-05T11:21:48.367781+00:00","updated_at":"2026-07-05T11:21:48.367781+00:00"}