Graph-Constrained Policy Learning for Extreme Clinical Code Prediction

• URL: http://arxiv.org/abs/2607.11954v1
• Date: Sat, 11 Jul 2026 23:10:01 GMT
• Title: Graph-Constrained Policy Learning for Extreme Clinical Code Prediction
• Abstract summary: Clinical code prediction maps unstructured discharge summaries to ICD-10-CM leaf codes in a large, sparse, and deeply hierarchical label space.<n>Most systems treat the task as flat multi-label classification, scoring codes independently and providing limited training signal for rare labels.<n>We propose a graph-constrained policy that formulates ICD prediction as a finite-horizon decision process over a pruned code hierarchy.
• Score: 1.116791976498254
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Clinical code prediction maps unstructured discharge summaries to ICD-10-CM leaf codes in a large, sparse, and deeply hierarchical label space. Most systems treat the task as flat multi-label classification, scoring codes independently and providing limited training signal for rare labels. We propose a graph-constrained traversal policy that formulates ICD prediction as a finite-horizon decision process over a pruned code hierarchy. A single language model descends the graph level by level, selecting valid child nodes until billable leaf codes are reached. This converts extreme multi-label prediction into sparse, hierarchy-aware subset decisions while guaranteeing structurally valid outputs. On MIMIC-IV discharge summaries, our best supervised policy, SFT-1+, achieves 0.709 micro-F1 on a curated 50-code subset and 0.527 micro-F1 on the full 15,761-code space, outperforming flat baselines including CAML, LAAT, and PLM-ICD. In the full setting, SFT-1+ improves over the strongest flat baseline by 0.044 micro-F1 and 0.157 macro-F1, suggesting that graph-constrained decomposition mitigates the rare-code bottleneck. A controlled factorial study evaluates architecture, training algorithm, and data budget. Across both scales, one shared policy matches a three-specialist cascade while avoiding its context-window overflow on 28-32% of full-space test notes. Increasing supervised trajectory data is the only intervention that consistently improves performance, while GRPO reinforcement learning provides no benefit over supervised continuation with matched data. These results show that simple graph-constrained policy learning can outperform more complex flat, cascaded, and reinforcement-learning alternatives for extreme clinical code prediction.

Related papers

• Accurate and Resource-Efficient Federated Continual Learning [3.983545455427237]
  Federated continual learning (FCL) must learn from distributed task streams under limited resources.<n>We propose FedRAN, a resource-aware analytic FCL framework that replaces gradient-based updates with compact random feature statistics.<n>FedRAN improves average accuracy by up to 4.8 percentage points over the strongest baseline, uses 30.6-121.8$times$ less per-client communication than optimization-based FCL, and is 190.3$times$ faster on average than gradient-based baselines.
  arXiv  Detail & Related papers  (2026-06-09T22:14:16Z)
• Verifier-Free RL for LLMs via Intrinsic Gradient-Norm Reward [69.99652051809737]
  In this work, we propose Verifier-free Intrinsic Gradient-Norm Reward (VIGOR)<n>VIGOR is a simple reward that uses only the policy model itself.<n>It exhibits cross-domain transfer to code benchmarks when trained only on math data.
  arXiv  Detail & Related papers  (2026-05-11T03:15:37Z)
• Learning Sparse Label Couplings for Multilabel Chest X-Ray Diagnosis [0.5735035463793009]
  We study multilabel classification of chest X-rays and present a simple, strong pipeline built on SE-ResNeXt101 $(32 times 4d)$.<n>The backbone is finetuned for 14 thoracic findings with a sigmoid head, trained using Multilabel Iterative Stratification (MIS) for robust cross-validation.<n>We propose a lightweight Label-Graph Refinement module placed after the classifier: given per-label probabilities, it learns a sparse, trainable inter-label coupling matrix.
  arXiv  Detail & Related papers  (2025-11-11T03:38:01Z)
• Federated Learning for ICD Classification with Lightweight Models and Pretrained Embeddings [0.9668407688201359]
  This study investigates the feasibility and performance of federated learning for multi-label ICD code classification.<n>We propose a scalable pipeline combining frozen text embeddings with simple multilayer perceptron (MLP) classifiers.
  arXiv  Detail & Related papers  (2025-07-03T18:58:36Z)
• Class Balance Matters to Active Class-Incremental Learning [61.11786214164405]
  We aim to start from a pool of large-scale unlabeled data and then annotate the most informative samples for incremental learning.<n>We propose Class-Balanced Selection (CBS) strategy to achieve both class balance and informativeness in chosen samples.<n>Our CBS can be plugged and played into those CIL methods which are based on pretrained models with prompts tunning technique.
  arXiv  Detail & Related papers  (2024-12-09T16:37:27Z)
• Continual Learning on a Diet: Learning from Sparsely Labeled Streams Under Constrained Computation [123.4883806344334]
  We study a realistic Continual Learning setting where learning algorithms are granted a restricted computational budget per time step while training. We apply this setting to large-scale semi-supervised Continual Learning scenarios with sparse label rates. Our extensive analysis and ablations demonstrate that DietCL is stable under a full spectrum of label sparsity, computational budget, and various other ablations.
  arXiv  Detail & Related papers  (2024-04-19T10:10:39Z)
• Automated clinical coding using off-the-shelf large language models [10.365958121087305]
  The task of assigning diagnostic ICD codes to patient hospital admissions is typically performed by expert human coders. Efforts towards automated ICD coding are dominated by supervised deep learning models. In this work, we leverage off-the-shelf pre-trained generative large language models to develop a practical solution.
  arXiv  Detail & Related papers  (2023-10-10T11:56:48Z)
• M-Tuning: Prompt Tuning with Mitigated Label Bias in Open-Set Scenarios [58.617025733655005]
  We propose a vision-language prompt tuning method with mitigated label bias (M-Tuning)<n>It introduces open words from the WordNet to extend the range of words forming the prompt texts from only closed-set label words to more, and thus prompts are tuned in a simulated open-set scenario.<n>Our method achieves the best performance on datasets with various scales, and extensive ablation studies also validate its effectiveness.
  arXiv  Detail & Related papers  (2023-03-09T09:05:47Z)
• Improved techniques for deterministic l2 robustness [63.34032156196848]
  Training convolutional neural networks (CNNs) with a strict 1-Lipschitz constraint under the $l_2$ norm is useful for adversarial robustness, interpretable gradients and stable training. We introduce a procedure to certify robustness of 1-Lipschitz CNNs by replacing the last linear layer with a 1-hidden layer. We significantly advance the state-of-the-art for standard and provable robust accuracies on CIFAR-10 and CIFAR-100.
  arXiv  Detail & Related papers  (2022-11-15T19:10:12Z)
• Code Compliance Assessment as a Learning Problem [0.15229257192293197]
  We formulate code compliance assessment as a machine learning (ML) problem. We generate a prediction on the code's compliance, non-compliance, or irrelevance. Our tool Policy2Code achieves classification accuracies of (59%, 71%) and search MRR of (0.05, 0.21) In a user study, 24% Policy2Code detections were accepted compared to 7% for CodeBERT.
  arXiv  Detail & Related papers  (2022-09-10T05:41:04Z)
• Learning, compression, and leakage: Minimising classification error via meta-universal compression principles [87.054014983402]
  A promising group of compression techniques for learning scenarios is normalised maximum likelihood (NML) coding. Here we consider a NML-based decision strategy for supervised classification problems, and show that it attains PAC learning when applied to a wide variety of models. We show that the misclassification rate of our method is upper bounded by the maximal leakage, a recently proposed metric to quantify the potential of data leakage in privacy-sensitive scenarios.
  arXiv  Detail & Related papers  (2020-10-14T20:03:58Z)
• Reliable Label Bootstrapping for Semi-Supervised Learning [19.841733658911767]
  ReLaB is an unsupervised preprossessing algorithm which improves the performance of semi-supervised algorithms in extremely low supervision settings. We show that the selection of the network architecture and the self-supervised algorithm are important factors to achieve successful label propagation. We reach average error rates of $boldsymbol22.34$ with 1 random labeled sample per class on CIFAR-10 and lower this error to $boldsymbol8.46$ when the labeled sample in each class is highly representative.
  arXiv  Detail & Related papers  (2020-07-23T08:51:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.