Machine Learning News

AsymVerify, a confidence-gated verification system for political evasion detection, scored 0.85 Macro F1 on the evaluation split (D_eval, n=237), placing 2nd out of 41 teams at SemEval-2026 Task 6. The system, developed …

Researchers propose Graph Wavelet Compressed Sensing (GWCS), a learning-based framework for offline compression of graph signals that combines a nonparametric multilevel importance sampler with a scale-aware graph neural…

Researchers propose Uncertainty-Guided Cascade Forward Refinement (U-CFR), a new inference-time framework for interactive image segmentation that enables models to autonomously self-correct after each user interaction. U…

Researchers introduce StressGAT, a Graph Attention Network that achieves 88.62% accuracy in stress recognition by using Differential Action Units to normalize individual facial expressions relative to neutral baselines. …

Researchers introduced Lumera, a benchmark and pipeline for engine-native, light-aware 3D scene parsing from a single image, built from 2,513 UE5 projects with 3.73M components and 102.6K parametric lights. Lumera-Box ac…

Researchers at NVIDIA have enhanced preconditioned gradient methods like SOAP and Muon to overcome computational cost and numerical stability challenges in large-scale LLM pretraining, demonstrating that these optimizers…

Researchers introduced CvAdamW, a variant of the AdamW optimizer that monitors attention logit variance as specific heat to detect and accelerate grokking in neural networks, enabling generalization at epoch 2802 in a 40…

A new study from arXiv identifies Spectral Drift in neural network internal activations as a reliable indicator of misclassifications, with failures showing a 1.9% increase in drift (p<0.001). The researchers introduce S…

A new study tests causal emergence in an active inference agent whose architecture separates a fast perception latent z from a slow global latent g, finding that the integrated information measure Φr concentrates in g an…

Researchers propose Cardinality-Decomposed Loss (CDL) to address a silent failure in Graph Neural Networks for heterogeneous recommendation graphs, where Bayesian Personalized Ranking (BPR) causes attribute embeddings to…

A study of Microsoft's Aurora weather forecasting foundation model fine-tuned for atmospheric chemistry finds that while it captures a first-order ozone response to reactive nitrogen, it does not enforce the chemical con…

A new robust Q-learning algorithm, BR-Async-Q, achieves the first robustness guarantee for asynchronous Q-learning under both reward and state corruption, matching vanilla Q-learning error bounds up to a small additive t…

Researchers propose Hierarchical Implicit Q-Chunking (HiQC), an offline goal-conditioned reinforcement learning algorithm that combines high-level latent planning with low-level action chunking to address the curse of ho…

A research-led guide compares top multimodal embedding models for image-text retrieval, visual documents, video, audio, and multimodal RAG, emphasizing that retrieval systems must preserve fine-grained details lost durin…

A developer has built Bio-Tuning Glasses, an experimental invisible biofeedback interface that uses edge AI and adaptive optics to subtly modify a user's environment based on physiological state, rather than sending noti…

XLA, the compiler under JAX, TensorFlow, and PyTorch/XLA, uses two intermediate representations: classic HLO (a hand-built C++ IR) and MLIR dialects such as StableHLO and CHLO, with a translation layer moving programs be…

Researchers upgraded a deep learning model using 1960s CORONA satellite imagery to automatically detect archaeological sites in Mesopotamia, achieving over 85% Intersection over Union and 90% detection accuracy in the Ab…

Researchers have developed a Deformable Detection Transformer (DETR)-based object detection model that achieved up to 28-37% improvement over prior work on COCO-style metrics for automated cuneiform sign detection, using…

A new deep learning model called Q2000, developed by researchers using G-LiHT Lidar data, detects ancient Maya settlements across 35,584 km2 of the Yucatan Peninsula with an F1 score of 0.89, matching the accuracy of sma…

A new paper on arXiv proposes 'computational identifiability,' a framework that replaces theoretical, asymptotic identification conditions with a finite computational search procedure for an empirical estimator within a …