arXiv:2604.24877v1 Announce Type: cross Abstract: Controlling illumination in images is essential for photography and visual content creation. While closed-source models have demonstrated impressive illumination control, open-source alternatives either require heavy control inputs like depth maps or do not release their data and code. We present a fully open-source and reproducible pipeline for learning illumination control […]
BenchGuard: Who Guards the Benchmarks? Automated Auditing of LLM Agent Benchmarks
arXiv:2604.24955v1 Announce Type: cross Abstract: As benchmarks grow in complexity, many apparent agent failures are not failures of the agent at all – they are failures of the benchmark itself: broken specifications, implicit assumptions, and rigid evaluation scripts that penalize valid alternative approaches. We propose employing frontier LLMs as systematic auditors of evaluation infrastructure, and […]
Loop Corrections to the Training Error and Generalization Gap of Random Feature Models
arXiv:2604.12827v3 Announce Type: replace-cross Abstract: We investigate random feature models in which neural networks sampled from a prescribed initialization ensemble are frozen and used as random features, with only the readout weights optimized. Adopting a statistical-physics viewpoint, we study the training error, test error, and generalization gap beyond the mean kernel approximation. Since the predictor […]
Nautile-370M: Spectral Memory Meets Attention in a Small Reasoning Model
arXiv:2604.24809v1 Announce Type: cross Abstract: We present Nautile-370M, a 371-million-parameter small language model designed for efficient reasoning under strict parameter and inference budgets. Nautile-370M uses a hybrid backbone in which two SeqCond Attention (SCA) layers, a linear-time spectral sequence operator inspired by SeqCondenser, alternate with one transformer layer. This design aims to retain the long-context […]
V.O.I.C.E (Voice, Ownership, Identity, Control, Expression): Risk Taxonomy of Synthetic Voice Generation From Empirical Data
arXiv:2604.24794v1 Announce Type: cross Abstract: As generative voice models are rapidly advancing in both capabilities and public utilization, the unconsented collection, reuse, and synthesis of voice data are introducing new classes of privacy, security and governance risk that are poorly captured by existing, largely uniform threat models. To fill the gap, we present V.O.I.C.E, a […]
From Prototype to Classroom: An Intelligent Tutoring System for Quantum Education
arXiv:2604.24807v1 Announce Type: cross Abstract: Quantum computing instructors face a compounding problem: the concepts are counterintuitive, the mathematical formalism is dense, and qualified faculty are scarce outside a small number of well-resourced institutions. Our prior work introduced a knowledge-graph-augmented tutoring prototype with two specialized LLM agents: a Teaching Agent for dynamic interaction and a Lesson […]
A Comparative Evaluation of AI Agent Security Guardrails
arXiv:2604.24826v1 Announce Type: cross Abstract: This report presents a comparative evaluation of DKnownAI Guard in AI agent security scenarios, benchmarked against three competing products: AWS Bedrock Guardrails, Azure Content Safety, and Lakera Guard. Using human annotation as the ground truth, we assess each guardrail’s ability to detect two categories of risks: threats to the agent […]
Learning biophysical models of gene regulation with probability flow matching
arXiv:2604.25062v1 Announce Type: new Abstract: Cellular differentiation is governed by gene regulatory networks, the high-dimensional stochastic biochemical systems that determine the transcriptional landscape and mediate cellular responses to signals and perturbations. Although single-cell RNA sequencing provides quantitative snapshots of the transcriptome, current methods for inferring gene-regulatory dynamics often lack mechanistic interpretability and fail to generalize […]
SIEVES: Selective Prediction Generalizes through Visual Evidence Scoring
arXiv:2604.25855v1 Announce Type: cross Abstract: Multimodal large language models (MLLMs) achieve ever-stronger performance on visual-language tasks. Even as traditional visual question answering benchmarks approach saturation, reliable deployment requires satisfying low error tolerances in real-world out-of-distribution (OOD) scenarios. Precisely, selective prediction aims to improve coverage, i.e. the share of inputs the system answers, while adhering to […]
Learning from Noisy Preferences: A Semi-Supervised Learning Approach to Direct Preference Optimization
arXiv:2604.24952v1 Announce Type: cross Abstract: Human visual preferences are inherently multi-dimensional, encompassing aesthetics, detail fidelity, and semantic alignment. However, existing datasets provide only single, holistic annotations, resulting in severe label noise: images that excel in some dimensions but are deficient in others are simply marked as winner or loser. We theoretically demonstrate that compressing multi-dimensional […]
asRoBallet: Closing the Sim2Real Gap via Friction-Aware Reinforcement Learning for Underactuated Spherical Dynamics
arXiv:2604.24916v1 Announce Type: cross Abstract: We introduce asRoBallet, to the best of our knowledge, the first successful deployment of reinforcement learning (RL) on a humanoid ballbot hardware. Historically, ballbots have served as a canonical benchmark for underactuated and nonholonomic control, which are characterized by a reality gap in complex friction models for wheel-sphere-ground interactions. While […]
Transformer Approximations from ReLUs
arXiv:2604.24878v1 Announce Type: cross Abstract: We provide a systematic recipe for translating ReLU approximation results to softmax attention mechanism. This recipe covers many common approximation targets. Importantly, it yields target-specific, economic resource bounds beyond universal approximation statements. We showcase the recipe on multiplication, reciprocal computation, and min/max primitives. These results provide new analytical tools for […]