{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/arxiv-2606-21447","title":"Precision Recall Controllable Radiology Report Generation via Hybrid Natural Language and Clinical Reward Learning","arxiv_id":"2606.21447","date":"2026-06-19","proceeding":null,"authors":["Ling Chen","Ruinan Jin","Jun Luo","Hanliang Chen","Quirin Strotzer","Rongkai Yan","Yuan Xue","Luciano Prevedello","Dufan Wu"],"abstract":"Automated radiology report generation (RRG) has gained increasing attention because it can reduce the heavy workload of clinical report writing. However, most existing methods mainly optimize for natural language generation (NLG) metrics that focus on language fluency, while providing little control over clinically important factors such as precision and recall. As consequence, generated reports may be fluent but not well aligned with different clinical needs. To address this challenge, we propose a reinforcement learning framework for precision recall controllable RRG, where a control parameter explicitly adjusts the trade-off between clinical precision and recall during inference. This design allows the model to flexibly generate reports according to different clinical requirements. To ensure clinical correctness, we introduce a clinical reward into the training objective, which helps improve clinical efficacy (CE) beyond standard language-based optimization. In addition, we apply a group-relative training strategy that normalizes rewards within each training group, reducing reward variance and improving training stability. Extensive experiments on the MIMIC-CXR dataset show that our method consistently outperforms state-of-the-art approaches in both NLG and CE evaluation metrics, while providing reliable control over the CE precision recall trade-off.","url_abs":"https://arxiv.org/abs/2606.21447","url_pdf":"https://arxiv.org/pdf/2606.21447","source":{"archive":null,"snapshot":"2025-07-28","note":"not in the Papers with Code archive (frozen at the snapshot)","row_kind":"graph","title_abstract_authors_date":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)"},"code_links":[],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2606.21447","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2606.21447"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"mentioned_in_github":null,"is_official":null,"provenance":"deterministic:regex_extraction","mentioned_in_paper":null,"url":"https://github.com/98lingchen/MICCAI2026","reach":null}],"summary":{"ran":7,"ran_draft_wrong":3,"unverified":3},"by_repo_kind":{"found_in_text":{"samples":13,"ran":10,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":13,"samples":[{"code_sha256_prefix":"cab04f1e2101c79a","entry":"array_to_str","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/rewards.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/rewards.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"cab04f1e2101c79a"}},{"code_sha256_prefix":"83045338e170467c","entry":"attention","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/encoder_decoder.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/encoder_decoder.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"83045338e170467c"}},{"code_sha256_prefix":"3cbefcc988ce7b35","entry":"build_lr_scheduler","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/optimizers.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/optimizers.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3cbefcc988ce7b35"}},{"code_sha256_prefix":"3d5d349dca69b884","entry":"build_optimizer","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/optimizers.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/optimizers.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"3d5d349dca69b884"}},{"code_sha256_prefix":"6cdff690e29fd5e2","entry":"clones","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/encoder_decoder.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/encoder_decoder.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6cdff690e29fd5e2"}},{"code_sha256_prefix":"0bc1f8b3524f6b1b","entry":"compute_loss","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/loss.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/loss.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0bc1f8b3524f6b1b"}},{"code_sha256_prefix":"dbc167454f546647","entry":"get_lr","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/optimizers.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/optimizers.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dbc167454f546647"}},{"code_sha256_prefix":"bfac58a04b6835f5","entry":"pad_unsort_packed_sequence","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/att_model.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/att_model.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"bfac58a04b6835f5"}},{"code_sha256_prefix":"86839e01585b85d8","entry":"sort_pack_padded_sequence","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/att_model.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/att_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"86839e01585b85d8"}},{"code_sha256_prefix":"86e3baa66a006e69","entry":"subsequent_mask","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/encoder_decoder.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/encoder_decoder.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"86e3baa66a006e69"}},{"code_sha256_prefix":"dadd98c5d250ffbd","entry":"get_absolute_score","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/rewards.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/rewards.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dadd98c5d250ffbd"}},{"code_sha256_prefix":"8387de6a9eff60d4","entry":"get_self_critical_reward","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/rewards.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/rewards.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8387de6a9eff60d4"}},{"code_sha256_prefix":"40b2937c7e20170d","entry":"pack_wrapper","repo":"98lingchen/MICCAI2026","repo_kind":"found_in_text","path":"modules/att_model.py","file_url":"https://github.com/98lingchen/MICCAI2026/blob/HEAD/modules/att_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"40b2937c7e20170d"}}]},"arxiv_metadata":{"licence":"arXiv metadata, CC0 1.0 (https://info.arxiv.org/help/license)","fields":["title","abstract","authors","date"],"primary_category":"cs.CL","source":"arxiv_2026.jsonl"},"syntology_extracted_results":null}