New

Tasks involving normative knowledge include concept interpretation, eligibility determination, procedure explanation, material requirements, and service-outcome assessmen

Deploying a large language model (LLM) requires choosing the right accelerator, parallelization strategy, engine configuration, and replica count amongst ~140 other varia

Vision-language-action (VLA), video-action (VA), and world action models (WAMs) form the backbone of modern physical AI systems, yet their real-time single-batch inferenc

LLMs are becoming an indispensable part of network control systems due to their impressive reasoning. Before such LLM-enabled systems can be deployed, it is imperative to

Pairwise relevance judgments constrain both the loss numerator and the ideal-ranking denominator of nDCG. We characterize the exact worst-case loss of replacing a ranking

A low-precision agent can call its full-precision (BF16) model at selected decisions as a fallback. A learned precision router chooses where to call it from the disagreem

Before a new policy replaces a deployed one, we often want a certificate, computed from logged episodes, that it is better. We ask how many episodes such a certificate ne

Training large language models directly in 4-bit floating-point (FP) formats promises substantial improvements in throughput and energy efficiency. While some recipe desi

Uniform convergence asks how many random samples suffice for empirical averages to approximate expectations simultaneously over an entire class of functions. It is a cent

Evaluating a fine-tuned coaching model with expert graders is slow and expensive, and an LLM judge cannot be trusted without ground truth. We present a staged pipeline th

Sampling temperature can substantially reshape autoregressive generation, but it is unclear when temperature changes in long-range correlations reflect a genuine phase tr

We study realizable multiclass PAC learning with query feedback, where each training example provides a single binary response indicating whether its true label belongs t

Advances in reasoning, coding, and tool use enable LLM agents to solve operations research (OR) problems through iterative problem modeling, program construction, and ver

Domain ontologies require review against knowledge recorded in evolving external resources. We propose COATER, a method that manages executable review components called a

A universal forecaster is trained once and applied everywhere, so it must serve tasks whose schemas, forecast horizons, and temporal resolutions all differ. The number of

LLM-as-a-Judge systems are increasingly employed for automated evaluation, yet remain susceptible to systematic biases that undermine their reliability. We demonstrate th

Probabilistic circuits (PCs) are generative models that support efficient probabilistic inference. Adding latent states to a PC increases its representational capacity, b

Sepsis is a major public health concern, and many sepsis patients die due to ineffective hemodynamic management. Artificial intelligence could support hemodynamic decisio

AI agents increasingly interact with external environments to complete complex user tasks. These interactions can lead to unexpected behavior and expose agents to securit

Location encoders learn compact representations of latitude--longitude coordinates for downstream geospatial prediction tasks. Existing approaches are pretrained using co

We study the Muon optimizer using sharpness and provide the first empirical evidence that the Flat Minima Hypothesis also applies to Muon. Further, we derive Muon's impli

Recent studies have proposed reinforcement learning with verifiable rewards (RLVR) methods that enable LLMs to generate high-quality RTL circuits from natural-language sp

Large Language Models (LLMs) are presently well assessed by expert-level mathematics and physics benchmarks which focus on theoretical, symbolic reasoning. However, we la

We present Torch-DPhys, a Torch-native framework implemented in PyTorch for differentiable physics simulation and hybrid physics-machine learning. It implements selected

Low-light video enhancement requires the recovery of brightness, color, and fine spatial structures while preserving temporal coherence in the presence of noise and motio

Multi-agent LLM systems exchange messages before acting through tools, but message delivery, signing authority, and tool-side enforcement are separate trust boundaries. W

Until April 2026, the National Vulnerability Database labeled nearly every published Common Vulnerabilities and Exposures (CVE) entry with a Common Vulnerability Scoring

Tool-using agents are trained and evaluated almost exclusively with strong simulated users who state every detail clearly, yet real users are often weaker. We construct w

Neural networks whose inputs are functions, as in operator learning, are usually shown to be universal by existence arguments, either by reduction to finite-dimensional n

Learned representations often contain geometric structure induced by symmetries in the underlying data. We ask whether a known continuous parameter space can instead be u

Feedback-driven LLM optimizers are increasingly used for evolutionary search, Bayesian optimization, heuristic design, and scientific discovery. Yet rapid initial gains o

Pixel-based language models process rendered text without relying on a fixed subword vocabulary, offering a promising approach to noisy and multilingual inputs. However,

Large language model (LLM) agents now place options orders at retail brokers, and developers publish open-source options agents faster than anyone has examined them. Prio

Safety classifiers for large language models (LLMs) are typically built and evaluated on standard-script text, yet Bangla, like many lower-resource languages, is routinel

Search-augmented language models increasingly serve as long-horizon research agents, yet most open training methods assume a fixed user goal throughout search. We study m

Text-conditioned 3D human--object interaction (HOI) generation remains challenging because individually plausible human and object motions do not necessarily form a coher

Action-conditioned world models support planning and policy evaluation by predicting future states. However, learning an accurate one-step predictor can require substanti

Tabular foundation models (TFMs) are increasingly deployed through black-box APIs and trained on large collections of real-world tabular datasets. As this training paradi

Browser agents automate complex workflows by reading and acting on arbitrary page content, but this flexibility exposes them to prompt injection attacks and unintended da

Computational models have explanatory power to understand cognition in individuals, but there is usually not enough training data to build such models with traditional ma