Multidisciplinary team (MDT) consultations are the gold standard for cancer care decision-making, yet current practice lacks structured mechanisms for quantifying consensus and ensuring decision traceability. We introduce a Multi-Agent Medical Decision Consensus Matrix System that deploys seven specialized large language model agents, including an oncologist, a radiologist, a nurse, a psychologist, a patient advocate, a nutritionist and a rehabilitation therapist, to simulate realistic MDT workflows. The framework incorporates a mathematically grounded consensus matrix that uses Kendall's coefficient of concordance to objectively assess agreement. To further enhance treatment recommendation quality and consensus efficiency, the system integrates reinforcement learning methods, including Q-Learning, PPO and DQN. Evaluation across five medical benchmarks (MedQA, PubMedQA, DDXPlus, MedBullets and SymCat) shows substantial gains over existing approaches, achieving an average accuracy of 87.5% compared with 83.8% for the strongest baseline, a consensus achievement rate of 89.3% and a mean Kendall's W of 0.823. Expert reviewers rated the clinical appropriateness of system outputs at 8.9/10. The system guarantees full evidence traceability through mandatory citations of clinical guidelines and peer-reviewed literature, following GRADE principles. This work advances medical AI by providing structured consensus measurement, role-specialized multi-agent collaboration and evidence-based explainability to improve the quality and efficiency of clinical decision-making.

We present a new algorithm for solving optimization problems with objective functions that are the sum of a smooth function and a (potentially) nonsmooth regularization function, and nonlinear equality constraints. The algorithm may be viewed as an extension of the well-known proximal-gradient method that is applicable when constraints are not present. To account for nonlinear equality constraints, we combine a decomposition procedure for computing trial steps with an exact merit function for determining trial step acceptance. Under common assumptions, we show that both the proximal parameter and merit function parameter eventually remain fixed, and then prove a worst-case complexity result for the maximum number of iterations before an iterate satisfying approximate first-order optimality conditions for a given tolerance is computed. Our preliminary numerical results indicate that our approach has great promise, especially in terms of returning approximate solutions that are structured (e.g., sparse solutions when a one-norm regularizer is used).

Large language models have demonstrated remarkable capabilities across diverse reasoning tasks, yet their performance on algorithmic reasoning remains limited. To handle this limitation, we propose PRIME (Policy-Reinforced Iterative Multi-agent Execution), a framework comprising three specialized agents, an executor for step-by-step reasoning, a verifier for constraint checking, and a coordinator for backtracking control, optimized through group relative policy optimization. For comprehensive evaluation, we introduce PRIME-Bench, the largest algorithmic reasoning benchmark to date, comprising 86 tasks across 12 categories with 51,600 instances. Tasks span sorting algorithms, graph and tree structures, automata and state machines, symbolic reasoning, and constraint-based puzzles, with execution traces reaching over one million steps. Compared to baseline approach, PRIME improves average accuracy from 26.8% to 93.8%, a 250% relative gain. The largest improvements occur on tasks requiring sustained state tracking, with Turing machine simulation improving from 9% to 92% and long division from 16% to 94%. Ablation studies identify iterative verification as the primary contributor, preventing the error propagation that causes baseline approaches to fail catastrophically. Analysis across model scales (8B–120B parameters) reveals that smaller models benefit disproportionately, achieving accuracy comparable to models 8× larger.

Existing patent examination approaches face fundamental limitations: they struggle with comprehensive prior art coverage due to maximum similarity scoring without considering all claim elements, provide limited ranked retrieval through binary classification without confidence scoring, and incur substantial computational overhead while generating generic outputs that miss claim-specific details. To address these challenges, we introduce \textbf{Integrated Patent Prior Art Search with Claim-Aware Retrieval and Novelty Assessment} (IPAS-CARNA), a novel three-stage pipeline combining enhanced claim-document matching, continuous novelty assessment, and claim-aware summarization. Our approach models element-wise claim coverage through adaptive chunking and weighted aggregation, integrates continuous novelty scoring with confidence assessment, and introduces claim-aware summarization with dynamic length control. Extensive experiments on CLEF-IP 2013, USPTO examination records, and HUPD validation sets demonstrate significant improvements: MAP@100 of 0.342 with 14.8\% improvement in retrieval recall, 18.2\% improvement in NDCG@10 for novelty ranking, technical accuracy above 0.85, and ROUGE-L scores of 0.456 for summarization. Our work establishes an effective integrated solution for automated patent prior art analysis.

Neural network interpretability methods have produced powerful approaches for gradient-based feature attribution, energy landscape analysis, and uncertainty quantification. State-of-the-art methods rely on structural weight analysis, Monte Carlo dropout uncertainty estimation, and attention mechanisms for interpretable predictions with quantified confidence. However, existing methods face fundamental challenges: unreliable explanations with poor uncertainty quantification on complex medical imaging tasks, difficulty identifying important network weights due to fixed thresholds, and computational overhead from attention mechanisms operating without uncertainty guidance. We introduce Adaptive Uncertainty-Guided Interpretable Networks (AUGIN), a framework combining adaptive structural weight analysis, uncertainty-aware prediction intervals, and uncertainty-guided attention through three interconnected modules. Our approach models adaptive threshold computation evolving with layer depth and architecture, integrates feature-level uncertainty into prediction interval generation, and introduces uncertainty-guided attention focusing on uncertain regions. Experiments on ISLES stroke prediction, BraTS brain tumor segmentation, and CT-CTA thrombectomy outcome prediction demonstrate superior performance: 0.7891 AUC-ROC on ISLES (1.59 percentage point improvement), 0.8567 Dice score on BraTS, and 0.7456 attention IoU, while maintaining excellent uncertainty calibration (0.8123) and explanation fidelity (0.8234). Our work advances neural network interpretability by providing accurate predictions with reliable, uncertainty-quantified explanations for safety-critical medical imaging applications.

Structured pruning and quantization are fundamental techniques used to reduce the size of deep neural networks (DNNs) and typically are applied independently. Applying these techniques jointly via co-optimization has the potential to produce smaller, high-quality models. However, existing joint schemes are not widely used because of (1) engineering difficulties (complicated multi-stage processes), (2) black-box optimization (extensive hyperparameter tuning to control the overall compression), and (3) insufficient architecture generalization. To address these limitations, we present the framework GETA, which automatically and efficiently performs joint structured pruning and quantization-aware training on any DNNs. GETA introduces three key innovations: (i) a quantization-aware dependency graph (QADG) that constructs a pruning search space for generic quantization-aware DNN, (ii) a partially projected stochastic gradient method that guarantees layerwise bit constraints are satisfied, and (iii) a new joint learning strategy that incorporates interpretable relationships between pruning and quantization. We present numerical experiments on both convolutional neural networks and transformer architectures that show that our approach achieves competitive (often superior) performance compared to existing joint pruning and quantization methods.