Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
A developer proposes the term 'Automation Before Automation' (ABA) to describe an intermediate testing phase that involves automatically generating and executing exploratory, validation, and robustness tests before build…
Researchers introduced the Membership Inference Test (MINT) Demo 2, a framework to detect whether specific data were used in training machine learning models, achieving up to 90% accuracy in tests on face recognition mod…
Researchers propose SERAF, a semantics-enhanced retrieval-augmented time series forecasting framework that retrieves both historical time series segments and their self-generated textual descriptions to improve predictio…
Researchers developed Metric Match, a subset selection method that estimates LLM judge reliability from limited human annotations. The method achieved a win-rate of 0.838 against random selection across 15 datasets, redu…
Researchers introduced VigilFormer, a video anomaly detection framework combining deformable spatio-temporal attention with causal temporal modeling, achieving state-of-the-art AUC scores of 87.83% on UCF-Crime, 97.21% o…
Researchers from an undisclosed institution introduced Convolutional Nearest Neighbors (ConvNN), a unified framework that shows convolution and self-attention are special cases of k-nearest neighbor aggregation. ConvNN c…
Researchers introduced FairGen, a fairness-aware diffusion framework that synthesizes demographically balanced medical images while preserving pathology-relevant features. Applied to dermatology, radiology, and neuroimag…
Researchers propose Steady-Forcing, a memory and training framework for autoregressive video diffusion models that balances spatial persistence and motion continuity in long-horizon nature video generation. The method im…
Researchers introduced UtVAA, an ultra-tiny Vision Transformer architecture with a novel Affix Attention block for mobile image classification. The smallest variant has 204.67K parameters and 53.95M FLOPs, achieving comp…
Researchers propose Style-CCL, a multi-stage curriculum continual learning framework for content-preserving style transfer using Diffusion Transformers. The method trains a dual-branch SC-DiT from semantic to texture sty…
Researchers developed a high-throughput 3D behavioral phenotyping framework integrating deep learning object detection with binocular stereo vision for real-time monitoring of juvenile tilapia. The system automates body …
Researchers introduced X-Tokenizer, a multimodal action tokenizer for vision-language-action pretraining that uses Semantic Residual Quantization and Masked Action Modeling to create a discrete action language capturing …
Researchers introduced DR-DCI, a retriever-steered Direct Corpus Interaction framework that dynamically expands a local workspace for agentic search over large corpora. On Browsecomp-Plus, DR-DCI achieved 71.2% accuracy,…
Researchers introduced a foundation model-driven framework for cross-modal representation alignment between CT imaging and longitudinal EHR data, improving time-to-event prediction across tasks and institutions. Fusion s…
Researchers propose CogGuard, a proactive-warning framework for edge intelligent services that uses LLMs for offline profile construction and SLMs for online score prediction. The framework reduces profile construction t…
Researchers introduced PhoneHarness, a mixed-action benchmark and execution harness for phone-use agents that combines GUI, CLI, and tool actions. The system achieved a 75.0% pass rate on verifiable mobile workflows, out…
NVIDIA released Nemotron 3 Ultra, a 550B-parameter hybrid Mamba-Transformer model with 55B active parameters, achieving up to 6x higher inference throughput than state-of-the-art LLMs while maintaining accuracy. The mode…
Researchers achieved 98.91% accuracy in multimodal emotion recognition from physiological signals by combining LSTM, TCN, and Transformer models with late-fusion ensemble on the WESAD dataset. Transformer models excelled…
Researchers propose QPILOTS, a method that steers flow-matching and diffusion policies at inference time by projecting noisy intermediate states to clean action estimates for critic gradient computation. QPILOTS achieves…
Researchers introduced GRAPE, a training framework that progressively exposes neural network parameters to improve adversarial robustness. On CIFAR-10 with ResNet-18, GRAPE achieved 56.94% PGD-20 robust accuracy, a 5.24 …