arXiv:2510.14959v5 Announce Type: replace-cross Abstract: Reinforcement learning (RL), while powerful and expressive, can often prioritize performance at the expense of safety. Yet safety violations can lead to catastrophic outcomes in real-world deployments. Control Barrier Functions (CBFs) offer a principled method to enforce dynamic safety — traditionally deployed online via safety filters. While the result is […]
LUMINA: LLM-Guided GPU Architecture Exploration via Bottleneck Analysis
arXiv:2603.05904v2 Announce Type: replace-cross Abstract: GPU design space exploration (DSE) for modern AI workloads, such as Large-Language Model (LLM) inference, is challenging because of GPUs’ vast, multi-modal design spaces, high simulation costs, and complex design optimization objectives (e.g. performance, power and area trade-offs). Existing automated DSE methods are often prohibitively expensive, either requiring an excessive […]
GIFT: Reconciling Post-Training Objectives via Finite-Temperature Gibbs Initialization
arXiv:2601.09233v2 Announce Type: replace-cross Abstract: The prevailing post-training paradigm for Large Reasoning Models (LRMs) – Supervised Fine-Tuning (SFT) followed by Reinforcement Learning (RL) – suffers from an intrinsic optimization mismatch: the rigid supervision inherent in SFT induces distributional collapse, thereby exhausting the exploration space necessary for subsequent RL. In this paper, we reformulate SFT to […]
Unsupervised Symbolic Anomaly Detection
arXiv:2603.17575v1 Announce Type: cross Abstract: We propose SYRAN, an unsupervised anomaly detection method based on symbolic regression. Instead of encoding normal patterns in an opaque, high-dimensional model, our method learns an ensemble of human-readable equations that describe symbolic invariants: functions that are approximately constant on normal data. Deviations from these invariants yield anomaly scores, so […]
JAWS: Enhancing Long-term Rollout of Neural PDE Solvers via Spatially-Adaptive Jacobian Regularization
arXiv:2603.05538v2 Announce Type: replace-cross Abstract: Data-driven surrogate models can significantly accelerate the simulation of continuous dynamical systems, yet the step-wise accumulation of errors during autoregressive time-stepping often leads to spectral blow-up and unphysical divergence. Existing global regularization techniques can enforce contractive dynamics but uniformly damp high-frequency features, causing over-smoothing; meanwhile, long-horizon trajectory optimization methods are […]
CogGen: Cognitive-Load-Informed Fully Unsupervised Deep Generative Modeling for Compressively Sampled MRI Reconstruction
arXiv:2603.04438v2 Announce Type: replace-cross Abstract: Fully unsupervised deep generative modeling (FU-DGM) is promising for compressively sampled MRI (CS-MRI) when training data or compute are limited. Classical FU-DGMs such as DIP and INR rely on architectural priors, but the ill-conditioned inverse problem often demands many iterations and easily overfits measurement noise. We propose CogGen, a cognitive-load-informed […]
Resource Consumption Threats in Large Language Models
arXiv:2603.16068v2 Announce Type: replace-cross Abstract: Given limited and costly computational infrastructure, resource efficiency is a key requirement for large language models (LLMs). Efficient LLMs increase service capacity for providers and reduce latency and API costs for users. Recent resource consumption threats induce excessive generation, degrading model efficiency and harming both service availability and economic sustainability. […]
FoMo X: Modular Explainability Signals for Outlier Detection Foundation Models
arXiv:2603.17570v1 Announce Type: cross Abstract: Tabular foundation models, specifically Prior-Data Fitted Networks (PFNs), have revolutionized outlier detection (OD) by enabling unsupervised zero-shot adaptation to new datasets without training. However, despite their predictive power, these models typically function as opaque black boxes, outputting scalar outlier scores that lack the operational context required for safety-critical decision-making. Existing […]
Can Blindfolded LLMs Still Trade? An Anonymization-First Framework for Portfolio Optimization
arXiv:2603.17692v1 Announce Type: cross Abstract: For LLM trading agents to be genuinely trustworthy, they must demonstrate understanding of market dynamics rather than exploitation of memorized ticker associations. Building responsible multi-agent systems demands rigorous signal validation: proving that predictions reflect legitimate patterns, not pre-trained recall. We address two sources of spurious performance: memorization bias from ticker-specific […]
SWE-CI: Evaluating Agent Capabilities in Maintaining Codebases via Continuous Integration
arXiv:2603.03823v3 Announce Type: replace-cross Abstract: Large language model (LLM)-powered agents have demonstrated strong capabilities in automating software engineering tasks such as static bug fixing, as evidenced by benchmarks like SWE-bench. However, in the real world, the development of mature software is typically predicated on complex requirement changes and long-term feature iterations — a process that […]
RangeAD: Fast On-Model Anomaly Detection
arXiv:2603.17795v1 Announce Type: cross Abstract: In practice, machine learning methods commonly require anomaly detection (AD) to filter inputs or detect distributional shifts. Typically, this is implemented by running a separate AD model alongside the primary model. However, this separation ignores the fact that the primary model already encodes substantial information about the target distribution. In […]
FrescoDiffusion: 4K Image-to-Video with Prior-Regularized Tiled Diffusion
arXiv:2603.17555v1 Announce Type: cross Abstract: Diffusion-based image-to-video (I2V) models are increasingly effective, yet they struggle to scale to ultra-high-resolution inputs (e.g., 4K). Generating videos at the model’s native resolution often loses fine-grained structure, whereas high-resolution tiled denoising preserves local detail but breaks global layout consistency. This failure mode is particularly severe in the fresco animation […]