#machine-learning News & Analysis

TL++: Accuracy and Privacy Preserving Traversal Learning for Distributed Intelligent Systems

TL++ is a new distributed machine learning framework that enables training across isolated data sources while maintaining privacy and reducing communication overhead. The system uses secret-sharing techniques to protect sensitive activations while achieving superior accuracy compared to federated and split-learning baselines, demonstrating 13x communication reduction on CIFAR-10.

Gradient-based inverse lithography for EUV masks via the waveguide method and a physics-informed neural operator

Researchers present a novel gradient-based inverse lithography technology (ILT) for extreme ultraviolet (EUV) masks that uses physics-informed neural operators and automatic differentiation to optimize mask absorber permittivity. The method combines a differentiable waveguide approach with waveguide neural operators (WGNO) to recover mask structures achieving desired field patterns on wafers, demonstrated on realistic 2D and 3D absorbers at 11.2 nm wavelengths.

Explainable Control Framework (XCF) based on Fuzzy Model-Agnostic Explanation and LLM Agent-Supported Interface

Researchers propose an Explainable Control Framework (XCF) that uses fuzzy logic and large language models to make complex automated controllers transparent and understandable to humans. The system generates natural language explanations of controller decisions across multiple levels of abstraction, demonstrated through robotic control applications like inverted pendulums and obstacle avoidance.

Pulmonary Embolism Risk Stratification from CTPA and Medical Records: Vascular Graphs Are Not All You Need

A research study challenges the assumption that vascular graph neural networks improve pulmonary embolism risk stratification, finding that medical records and cardiac biomarkers alone outperform complex graph-based approaches. The findings suggest that sophisticated deep learning models may not capture clinically relevant information from vascular imaging data for this application.

SE-AGCNet: An End-to-End Framework for Joint Speech Enhancement and Loudness Control in Meeting Scenarios

Researchers propose SE-AGCNet, an end-to-end framework that jointly optimizes speech enhancement and automatic gain control for meeting scenarios. The approach addresses limitations of traditional discrete audio processing pipelines by leveraging synergy between the two tasks, improving speech quality, loudness consistency, and automatic speech recognition accuracy.

FORCE: Efficient VLA Reinforcement Fine-Tuning via Value-Calibrated Warm-up and Self-Distillation

Researchers introduce FORCE, a three-stage reinforcement learning framework that significantly improves the efficiency of fine-tuning Vision-Language-Action models for robotics. By addressing Q-function instability and low-quality exploration data, FORCE achieves 79% absolute improvement in success rates while reducing training time by 32.5%, eliminating the need for human intervention during deployment.

A welding penetration prediction model for laser welding process based on self-supervised learning using physics-informed neural networks

Researchers introduce SimPhysNet, a self-supervised learning algorithm that predicts laser welding penetration with 96.06% accuracy using only 200 labeled images—roughly 5% of typical datasets. The physics-informed neural network approach combines contrastive learning with few-shot learning to overcome the industrial manufacturing challenge of requiring extensive labeled data for quality assurance.

A cross-process welding penetration status prediction algorithm based on unsupervised domain adaptation in laser and TIG welding

Researchers have developed an unsupervised domain adaptation framework that enables deep learning models to predict weld penetration status across different welding processes without extensive relabeling. The approach achieves 80-81% accuracy in cross-process transfer between TIG and laser welding, significantly outperforming supervised baselines and reducing the cost of deploying AI systems to new welding environments.

On-Policy Self-Distillation with Sampled Demonstrations Reduces Output Diversity

Researchers reveal that on-policy self-distillation, a technique that improves single-model accuracy by using correct demonstrations as conditioning, reduces output diversity and flattens pass@k curves—meaning additional rollouts fail to boost performance. The method amplifies existing model biases rather than preserving probability ratios like optimal reinforcement learning does, causing models to concentrate on dominant modes and fail in out-of-distribution settings.

Towards Understanding The Calibration Benefits of Sharpness-Aware Minimization

Researchers demonstrate that Sharpness-Aware Minimization (SAM), a recently proposed neural network training method, significantly improves model calibration by reducing overconfidence in predictions. The study includes a new variant called CSAM that further enhances calibration performance across multiple datasets, with important implications for safety-critical AI applications.

HiT-JEPA: A Hierarchical Self-supervised Trajectory Embedding Framework for Similarity Computation

Researchers introduce HiT-JEPA, a hierarchical self-supervised learning framework that represents urban trajectory data across multiple semantic levels to improve similarity computation. The model captures fine-grained movement details, intermediate patterns, and high-level abstractions simultaneously, addressing limitations in existing approaches that struggle to balance local nuances with global dependencies.

Why Pool When You Can Flow? Active Learning with GFlowNets

Researchers introduce BALD-GFlowNet, a generative active learning framework that replaces traditional pool-based sample selection with generative sampling to dramatically improve scalability. The method maintains comparable performance to standard BALD while reducing computational costs independent of unlabeled dataset size, particularly valuable for drug discovery applications involving billions of molecular candidates.

Distribution Preference Optimization: A Fine-grained Perspective for LLM Unlearning

Researchers introduce DiPO (Distribution Preference Optimization), a novel algorithm for LLM unlearning that operates at the token distribution level rather than full response level. The method addresses limitations in existing approaches like NPO by constructing preference signals through selective amplification of model logits, achieving superior performance on benchmark tests while maintaining model utility.

CustomX: Unified Character, Action, and Scene Customization in Video World Models

CustomX is a new video world model that enables users to control multiple characters performing diverse actions within 3D environments using natural language prompts. The system combines realistic static scene generation with controllable character behaviors, synthesizing temporally coherent video clips while maintaining visual fidelity and character consistency.

Surrogate models for Rock-Fluid Interaction: A Grid-Size-Invariant Approach

Researchers develop grid-size-invariant neural network surrogate models for predicting rock-fluid interactions in porous media, offering a computationally cheaper alternative to traditional high-fidelity simulations. The approach demonstrates that UNet++ architecture outperforms standard UNet for this application, enabling significant memory reduction during training while maintaining prediction accuracy.

Incremental Residual Reinforcement Learning Toward Real-World Learning for Social Navigation

Researchers propose Incremental Residual Reinforcement Learning (IRRL), a new method that enables mobile robots to learn social navigation directly in physical environments without requiring large computational resources or replay buffers. The approach combines incremental learning with residual reinforcement learning to improve efficiency, achieving performance comparable to traditional methods while enabling real-world adaptation.

Getting past the pilot: Why so many AI test projects have trouble scaling

Business leaders from major corporations like Salesforce, Amgen, and Thomson Reuters are examining why AI pilot projects frequently fail to scale beyond initial testing phases. The analysis reveals critical gaps between proof-of-concept success and enterprise-wide deployment, with implications for how organizations approach AI implementation strategy.

Ensemble Feature Selection and Harris Hawks Optimization for Explainable Mental Health Risk Prediction in Female Sex Workers

Quant Convergence: Bridging Classical Value Investing and Modern Factor Models for Systematic Equity Selection

Legal Reasoning Is Not Lawyering: Rethinking Legal Benchmarks for Pro Se Access to Justice

Low-power analogue neural networks with trainable nonlinear connections for continuous control

Rapid FinFET Modelling Using an Autoencoder

Predicting Poets' Origins from Verse: A Computational Analysis of Regional Linguistic Fingerprints in the Complete Tang Poems

Lightweight Transformer Models for On-Device Fault Detection: A Benchmark Study on Resource-Constrained Deployment

Adaptive Machine Learning Framework for UAV Trajectory Optimization in O-RAN