CascadeDebate: Multi-Agent Deliberation for Cost-Aware LLM Cascades

Cascaded LLM systems coordinate models of varying sizes with human experts to balance accuracy, cost, and abstention under uncertainty. However, single-model tiers at each stage often struggle with ambiguous queries, triggering premature escalations to costlier models or experts due to under-confidence and inefficient compute scaling. CascadeDebate addresses this gap by inserting multi-agent deliberation directly at each tier's escalation boundary. Confidence-based routers activate lightweight agent ensembles only for uncertain cases, enabling consensus-driven resolution of ambiguities internally without invoking higher-cost upgrades. Our unified architecture alternates single-model inference with selective multi-agent deliberation across model scales, culminating in human experts as the final fallback. This design scales test-time compute dynamically according to query difficulty. Across five benchmarks spanning science, medicine, and general knowledge, CascadeDebate outperforms strong single-model cascades and standalone multi-agent systems by up to 26.75 percent. An online threshold optimizer proves essential, boosting accuracy by 20.98 to 52.33 percent relative improvement over fixed policies and enabling elastic adaptation to real-world distributions.

Privacy-Preserving Federated Fraud Detection in Payment Transactions with NVIDIA FLARE

Fraud-related financial losses continue to rise, while regulatory, privacy, and data-sovereignty constraints increasingly limit the feasibility of centralized fraud detection systems. Federated Learning (FL) has emerged as a promising paradigm for enabling collaborative model training across institutions without sharing raw transaction data. Yet, its practical effectiveness under realistic, non-IID financial data distributions remains insufficiently validated. In this work, we present a multi-institution, industry-oriented proof-of-concept study evaluating federated anomaly detection for payment transactions using the NVIDIA FLARE framework. We simulate a realistic federation of heterogeneous financial institutions, each observing distinct fraud typologies and operating under strict data isolation. Using a deep neural network trained via federated averaging (FedAvg), we demonstrate that federated models achieve a mean F1-score of 0.903 - substantially outperforming locally trained models (0.643) and closely approaching centralized training performance (0.925), while preserving full data sovereignty. We further analyze convergence behavior, showing that strong performance is achieved within 10 federated communication rounds, highlighting the operational viability of FL in latency- and cost-sensitive financial environments. To support deployment in regulated settings, we evaluate model interpretability using Shapley-based feature attribution and confirm that federated models rely on semantically coherent, domain-relevant decision signals. Finally, we incorporate sample-level differential privacy via DP-SGD and demonstrate favorable privacy-utility trade-offs...

Benchmarking State Space Models, Transformers, and Recurrent Networks for US Grid Forecasting

Selecting the right deep learning model for power grid forecasting is challenging, as performance heavily depends on the data available to the operator. This paper presents a comprehensive benchmark of five modern neural architectures: two state space models (PowerMamba, S-Mamba), two Transformers (iTransformer, PatchTST), and a traditional LSTM. We evaluate these models on hourly electricity demand across six diverse US power grids for forecast windows between 24 and 168 hours. To ensure a fair comparison, we adapt each model with specialized temporal processing and a modular layer that cleanly integrates weather covariates. Our results reveal that there is no single best model for all situations. When forecasting using only historical load, PatchTST and the state space models provide the highest accuracy. However, when explicit weather data is added to the inputs, the rankings reverse: iTransformer improves its accuracy three times more efficiently than PatchTST. By controlling for model size, we confirm that this advantage stems from the architecture's inherent ability to mix information across different variables. Extending our evaluation to solar generation, wind power, and wholesale prices further demonstrates that model rankings depend on the forecast task: PatchTST excels on highly rhythmic signals like solar, while state space models are better suited for the chaotic fluctuations of wind and price. Ultimately, this benchmark provides grid operators with actionable guidelines for selecting the optimal forecasting architecture based on their specific data environments.

Mi:dm 2.0 Korea-centric Bilingual Language Models

We introduce Mi:dm 2.0, a bilingual large language model (LLM) specifically engineered to advance Korea-centric AI. This model goes beyond Korean text processing by integrating the values, reasoning patterns, and commonsense knowledge inherent to Korean society, enabling nuanced understanding of cultural contexts, emotional subtleties, and real-world scenarios to generate reliable and culturally appropriate responses. To address limitations of existing LLMs, often caused by insufficient or low-quality Korean data and lack of cultural alignment, Mi:dm 2.0 emphasizes robust data quality through a comprehensive pipeline that includes proprietary data cleansing, high-quality synthetic data generation, strategic data mixing with curriculum learning, and a custom Korean-optimized tokenizer to improve efficiency and coverage. To realize this vision, we offer two complementary configurations: Mi:dm 2.0 Base (11.5B parameters), built with a depth-up scaling strategy for general-purpose use, and Mi:dm 2.0 Mini (2.3B parameters), optimized for resource-constrained environments and specialized tasks. Mi:dm 2.0 achieves state-of-the-art performance on Korean-specific benchmarks, with top-tier zero-shot results on KMMLU and strong internal evaluation results across language, humanities, and social science tasks. The Mi:dm 2.0 lineup is released under the MIT license to support extensive research and commercial use. By offering accessible and high-performance Korea-centric LLMs, KT aims to accelerate AI adoption across Korean industries, public services, and education, strengthen the Korean AI developer community, and lay the groundwork for the broader vision of K-intelligence. Our models are available at https://huggingface.co/K-intelligence. For technical inquiries, please contact midm-llm@kt.com.

Reliable Grid Forecasting: State Space Models for Safety-Critical Energy Systems

Accurate grid load forecasting is safety-critical: under-predictions risk supply shortfalls, while symmetric error metrics mask this operational asymmetry. We introduce a grid-specific evaluation framework (Asymmetric MAPE, Under-Prediction Rate, and Reserve Margin) that directly measures operational risk rather than statistical accuracy alone. Using this framework, we conduct a systematic evaluation of Mamba-based State Space Models for California grid forecasting on a weather-aligned CA ISO-TAC dataset spanning Nov 2023 to Nov 2025 (84,498 hourly records across 5 transmission areas). Our analysis reveals that standard accuracy metrics are poor proxies for operational safety: models with identical MAPE can require vastly different reserve margins. We demonstrate that forecast errors are weakly but statistically significantly associated with temperature (r = 0.16), motivating weather-aware modeling rather than loss function modification alone. The S-Mamba model achieves the lowest 99.5th-percentile reserve margin (14.12 percent) compared to 16.66 percent for iTransformer, demonstrating superior forecast reliability under a 99.5th-percentile tail-risk reserve proxy.

Transplant-Ready? Evaluating AI Lung Segmentation Models in Candidates with Severe Lung Disease

This study evaluates publicly available deep-learning based lung segmentation models in transplant-eligible patients to determine their performance across disease severity levels, pathology categories, and lung sides, and to identify limitations impacting their use in preoperative planning in lung transplantation. This retrospective study included 32 patients who underwent chest CT scans at Duke University Health System between 2017 and 2019 (total of 3,645 2D axial slices). Patients with standard axial CT scans were selected based on the presence of two or more lung pathologies of varying severity. Lung segmentation was performed using three previously developed deep learning models: Unet-R231, TotalSegmentator, MedSAM. Performance was assessed using quantitative metrics (volumetric similarity, Dice similarity coefficient, Hausdorff distance) and a qualitative measure (four-point clinical acceptability scale). Unet-R231 consistently outperformed TotalSegmentator and MedSAM in general, for different severity levels, and pathology categories (p<0.05). All models showed significant performance declines from mild to moderate-to-severe cases, particularly in volumetric similarity (p<0.05), without significant differences among lung sides or pathology types. Unet-R231 provided the most accurate automated lung segmentation among evaluated models with TotalSegmentator being a close second, though their performance declined significantly in moderate-to-severe cases, emphasizing the need for specialized model fine-tuning in severe pathology contexts.

MORE-CLEAR: Multimodal Offline Reinforcement learning for Clinical notes Leveraged Enhanced State Representation

Sepsis, a life-threatening inflammatory response to infection, causes organ dysfunction, making early detection and optimal management critical. Previous reinforcement learning (RL) approaches to sepsis management rely primarily on structured data, such as lab results or vital signs, and on a dearth of a comprehensive understanding of the patient's condition. In this work, we propose a Multimodal Offline REinforcement learning for Clinical notes Leveraged Enhanced stAte Representation (MORE-CLEAR) framework for sepsis control in intensive care units. MORE-CLEAR employs pre-trained large-scale language models (LLMs) to facilitate the extraction of rich semantic representations from clinical notes, preserving clinical context and improving patient state representation. Gated fusion and cross-modal attention allow dynamic weight adjustment in the context of time and the effective integration of multimodal data. Extensive cross-validation using two public (MIMIC-III and MIMIC-IV) and one private dataset demonstrates that MORE-CLEAR significantly improves estimated survival rate and policy performance compared to single-modal RL approaches. To our knowledge, this is the first to leverage LLM capabilities within a multimodal offline RL for better state representation in medical applications. This approach can potentially expedite the treatment and management of sepsis by enabling reinforcement learning models to propose enhanced actions based on a more comprehensive understanding of patient conditions.

GEMMAS: Graph-based Evaluation Metrics for Multi Agent Systems

Multi-agent systems built on language models have shown strong performance on collaborative reasoning tasks. However, existing evaluations focus only on the correctness of the final output, overlooking how inefficient communication and poor coordination contribute to redundant reasoning and higher computational costs. We introduce GEMMAS, a graph-based evaluation framework that analyzes the internal collaboration process by modeling agent interactions as a directed acyclic graph. To capture collaboration quality, we propose two process-level metrics: Information Diversity Score (IDS) to measure semantic variation in inter-agent messages, and Unnecessary Path Ratio (UPR) to quantify redundant reasoning paths. We evaluate GEMMAS across five benchmarks and highlight results on GSM8K, where systems with only a 2.1% difference in accuracy differ by 12.8% in IDS and 80% in UPR, revealing substantial variation in internal collaboration. These findings demonstrate that outcome-only metrics are insufficient for evaluating multi-agent performance and highlight the importance of process-level diagnostics in designing more interpretable and resource-efficient collaborative AI systems.

Improving Surgical Risk Prediction Through Integrating Automated Body Composition Analysis: a Retrospective Trial on Colectomy Surgery

Objective: To evaluate whether preoperative body composition metrics automatically extracted from CT scans can predict postoperative outcomes after colectomy, either alone or combined with clinical variables or existing risk predictors. Main outcomes and measures: The primary outcome was the predictive performance for 1-year all-cause mortality following colectomy. A Cox proportional hazards model with 1-year follow-up was used, and performance was evaluated using the concordance index (C-index) and Integrated Brier Score (IBS). Secondary outcomes included postoperative complications, unplanned readmission, blood transfusion, and severe infection, assessed using AUC and Brier Score from logistic regression. Odds ratios (OR) described associations between individual CT-derived body composition metrics and outcomes. Over 300 features were extracted from preoperative CTs across multiple vertebral levels, including skeletal muscle area, density, fat areas, and inter-tissue metrics. NSQIP scores were available for all surgeries after 2012.

Ground Reaction Force Estimation via Time-aware Knowledge Distillation

Human gait analysis with wearable sensors has been widely used in various applications, such as daily life healthcare, rehabilitation, physical therapy, and clinical diagnostics and monitoring. In particular, ground reaction force (GRF) provides critical information about how the body interacts with the ground during locomotion. Although instrumented treadmills have been widely used as the gold standard for measuring GRF during walking, their lack of portability and high cost make them impractical for many applications. As an alternative, low-cost, portable, wearable insole sensors have been utilized to measure GRF; however, these sensors are susceptible to noise and disturbance and are less accurate than treadmill measurements. To address these challenges, we propose a Time-aware Knowledge Distillation framework for GRF estimation from insole sensor data. This framework leverages similarity and temporal features within a mini-batch during the knowledge distillation process, effectively capturing the complementary relationships between features and the sequential properties of the target and input data. The performance of the lightweight models distilled through this framework was evaluated by comparing GRF estimations from insole sensor data against measurements from an instrumented treadmill. Empirical results demonstrated that Time-aware Knowledge Distillation outperforms current baselines in GRF estimation from wearable sensor data.