Researchers introduced two CUTLASS-based kernels that accelerate the SwiGLU activation function in large language models by up to 2.47x on NVIDIA H100 GPUs, shifting workloads from memory-bound to com…
Selective prediction is transforming large language models by enabling them to decide when to predict and when to defer to humans, significantly improving reliability in high-stakes environments. A ne…
Agentic AI systems are being explored to enhance Open RAN operations, offering explicit planning and self-management for network slice lifecycle and radio resource management. Research shows agentic p…
TabPack, a new ensemble of multilayer perceptrons for tabular data, minimizes hyperparameter tuning by sampling configurations in parallel and selecting the best members on the fly. It achieves compet…
Researchers developed WPG-MoE, a weak-prior-guided mixture-of-experts AI framework that personalizes depression detection by routing user expressions to specialized expert models, outperforming tradit…
A study using the Dublin EEG-based Reading Experiment Corpus (DERCo) found that the brain's response to words is influenced more by semantic relevance—how well a word fits into the surrounding context…
A new retrieval-augmented generation system, LongEval-RAG, outperforms complex alternatives by using a simple methodology, challenging the AI community's focus on intricate designs. The rule-minilm va…
Researchers introduced TR-RAG, a teacher-regularized retrieval-augmented generation method that improves multilingual language adherence and evidence-grounded correctness by using a teacher-student dy…
Researchers are using text-to-speech systems to create synthetic training data for spoken question answering in Luxembourgish, a low-resource language. By translating existing QA resources and synthes…
Researchers introduced Physics-Grounded Symbolic Architecture (PGSA), a new AI model that achieves exact linear identifiability across all physical regimes, overcoming the Gaussian limitation of Joint…
Researchers introduced TRACE, a new benchmark for evaluating conversational AI systems that use external tools, revealing that all current state-of-the-art evaluation frameworks fail to detect critica…
A new study on Principal Component Analysis (PCA) reveals convergence rates for handling multiple probability measures, with rates of n^{-1/2} + m^{-α} depending on embedding choice, and proves minima…
A new survey paper examines error limits and sample size constraints in operator learning using a minimax perspective, focusing on holomorphic operators and neural network approximations. The research…
Researchers have identified three core principles—divisibility, transferability, and stability—that govern compositional generalization in AI models, revealing that linear structures in neural represe…
Researchers have developed a new sampling framework that optimizes consistency models to solve inverse imaging problems more efficiently, outperforming traditional diffusion models on metrics like FID…
A new AI framework called EGAgent uses entity scene graphs to achieve long-term video understanding, reaching a 57.5% performance rate on the EgoLifeQA dataset. The technology aims to enable wearable …
Bayesian Invariant Prediction (BIP), a new probabilistic method for identifying stable features across environments, has been introduced to improve model generalization and scalability. The approach, …
Researchers propose using latent space generative world models to help autonomous vehicles handle unexpected situations, reducing reliance on vast training data. The approach integrates a neural netwo…
Researchers have developed a new adaptive learning rate approach for Stochastic Gradient Descent that adjusts rates based on objective estimates, achieving faster convergence in complex machine learni…
Researchers at the University of Texas at Austin introduced VSeek, a framework that uses reinforcement learning to transform long-video question answering into an interactive retrieval process, improv…