Machine learning news — deep learning, reinforcement learning, neural architecture search, diffusion models, and new ML frameworks and libraries.
Researchers propose MiLSD, a micro line-segment detector designed for resource-constrained devices such as microcontrollers, achieving competitive accuracy within a one-megabyte memory budget. The detector uses a compact…
Researchers propose D2PO, a framework that optimizes diffusion sampling policies by reformulating sampler optimization as a preference-based alignment problem using Direct Preference Optimization (DPO). D2PO introduces d…
Researchers propose WINRO, a training-free framework that improves text-to-motion alignment in diffusion models by selecting and refining initial noise tickets. The method enhances semantic consistency and temporal robus…
Researchers propose InductWave, a wavelet-based inductive embedding method for logical query answering on knowledge graphs that handles unseen entities during training. The model achieves state-of-the-art performance wit…
Researchers propose a multimodal solution for audio sentiment analysis that integrates audio and automatically generated multilingual text transcripts via cross-modal transformers. The approach improves sentiment classif…
Researchers propose a programmatic solution for generating product advertising headlines using reinforcement learning on transformer-based masked language models. The method conditions on multiple products and outperform…
Researchers propose PRoVeFL, a federated learning framework that combines privacy preservation, Byzantine robustness, and verifiable aggregation using multi-key fully homomorphic encryption across multiple servers. The f…
Researchers have developed BatteryMat, a three-tier machine-learning-to-density-functional-theory (ML-DFT) workflow for screening lithium-ion cathode materials, as detailed in a July 7, 2026 arXiv preprint. The pipeline …
Researchers at an undisclosed institution found that multi-teacher on-policy distillation for agentic language models can cause over-calling of tools, where models call tools on examples that should be answered directly.…
Researchers propose MILES, a framework that enables large language models to accumulate and reuse step-wise reasoning experience across sequential problems, improving accuracy and efficiency without requiring large-scale…
Researchers propose an evaluation framework combining regularized regression on item text embeddings with repeated cross-validated R-squared and two performance upper bounds—a reliability ceiling and a design ceiling—to …
Researchers developed a benchmark to evaluate vision-language models' ability to use spatial deictic expressions across four languages. Tests showed that current models use demonstratives like "this" and "that" different…
Researchers evaluated the relevance of several RAG metrics by comparing scores from four libraries (Ragas, DeepEval, RAGChecker, Opik) against human evaluators and standard metrics using a question-answering dataset. The…
Researchers propose R^3, a framework for rectifying textual violations in video advertisements while preserving original semantic intent. The system uses a group-relative compliance experience extractor and curriculum re…
Researchers developed SAMPA, a Whisper-based segmenter that transcribes Brazilian Portuguese speech while inserting markers for terminal prosodic boundaries. Fine-tuned on the NURC-SP dataset, SAMPA achieved competitive …
A new theoretical analysis of in-context search in large language models shows that when self-reflection reliably localizes early mistakes, sequential reasoning can yield exponential improvements over the base model, sol…
Researchers introduced MIRA-Math, a benchmark for evaluating mathematical reasoning where models must request a missing atomic fact under a strict budget before solving a problem. The benchmark contains 2,310 instances a…
Researchers introduced Pyligent, a training framework that teaches AI models to recover from errors by learning from failed reasoning branches. In tests on hidden graphs, Sudoku, and Blocksworld, Pyligent improved solve …
Researchers introduced a Siamese Neural Network framework to identify phase transitions in 3D percolation models using only 22 labeled probability points from non-critical regions. The method located percolation threshol…
Researchers at the GRAPES-3 experiment developed a graph neural network (GNN)-based pipeline for cosmic ray energy reconstruction, achieving improved accuracy and automated feature extraction on sparse detector arrays. T…