arXiv:2602.09130v1 Announce Type: new Abstract: Model compression is increasingly essential for deploying large language models (LLMs), yet existing evaluations are limited in method coverage and focus primarily on knowledge-centric benchmarks. Thus, we introduce UniComp, a unified evaluation framework for comparing pruning, quantization, and knowledge distillation. UniComp evaluates compressed models along three dimensions: performance, reliability, and efficiency, using a diverse set of capability- and safety-oriented benchmarks together with a hardware-aware efficiency analysis. Through extensive evaluation of six compression techniques […]

Ver mais

arXiv:2603.05724v1 Announce Type: new Abstract: The increasing intensity and frequency of wildfires are causing significant economic and societal impacts on communities through direct effects on the built environment, particularly critical infrastructure. Electrical systems can both initiate wild-fires (grid-to-fire) and be damaged by wildfire exposure (fire-to-grid). Therefore, resilient electric systems can both limit ignitions and be hardened such that they are more robust to fire demands. Researchers have investigated wildfire mitigation strategies using traditional transmission and distribution electrical test-system […]

Ver mais

arXiv:2605.00229v1 Announce Type: new Abstract: We study the problem of training diffusion and flow generative models to sample from target distributions defined by an exponential tilting of a base density; a formulation that subsumes both sampling from unnormalized densities and reward fine-tuning of pre-trained models. This problem can be approached from a stochastic optimal control (SOC) perspective, using adjoint-based or score matching methods, or from a non-equilibrium thermodynamics perspective. We provide a unified framework encompassing these approaches and […]

Ver mais

Health checks are a set of checks (duh) that you perform in order to tell whether an application/service is up, running & healthy or not. It’s usually one or more endpoints that reports the status, the response differs from language/framework to an other. Health checks are very useful especially when your application depends on other things like a database or even other services.

Ver mais

arXiv:2602.11247v1 Announce Type: new Abstract: Multi-turn prompt injection attacks distribute malicious intent across multiple conversation turns, exploiting the assumption that each turn is evaluated independently. While single-turn detection has been extensively studied, no published formula exists for aggregating per-turn pattern scores into a conversation-level risk score at the proxy layer — without invoking an LLM. We identify a fundamental flaw in the intuitive weighted-average approach: it converges to the per-turn score regardless of turn count, meaning a 20-turn […]

Ver mais

We consider the problem of learning the optimal policy for Markov decision processes with safety constraints. We formulate the problem in a reach-avoid setup. Our goal is to design online reinforcement learning algorithms that ensure safety constraints with arbitrarily high probability during the learning phase. To this end, we first propose an algorithm based on the optimism in the face of uncertainty (OFU) principle. Based on the first algorithm, we propose our main algorithm, which utilizes entropy regularization. […]

Ver mais