arXiv:2602.10971v1 Announce Type: cross Abstract: We consider the problem of heteroskedastic generalized linear bandits (GLBs) with adversarial corruptions, which subsumes various stochastic contextual bandit settings, including heteroskedastic linear bandits and logistic/Poisson bandits. We propose HCW-GLB-OMD, which consists of two components: an online mirror descent (OMD)-based estimator and Hessian-based confidence weights to achieve corruption robustness. This is computationally efficient in that it only requires ${O}(1)$ space and time complexity per iteration. Under the self-concordance assumption on the link function, […]

Ver mais

arXiv:2602.15252v1 Announce Type: new Abstract: In game theory, imperfect-recall decision problems model situations in which an agent forgets information it held before. They encompass games such as the “absentminded driver” and team games with limited communication. In this paper, we introduce the first benchmark suite for imperfect-recall decision problems. Our benchmarks capture a variety of problem types, including ones concerning privacy in AI systems that elicit sensitive information, and AI safety via testing of agents in simulation. Across […]

Ver mais

arXiv:2601.19940v1 Announce Type: new Abstract: Among hardware accelerators for deep-learning inference, data flow implementations offer low latency and high throughput capabilities. In these architectures, each neuron is mapped to a dedicated hardware unit, making them well-suited for field-programmable gate array (FPGA) implementation. Previous unrolled implementations mostly focus on fully connected networks because of their simplicity, although it is well known that convolutional neural networks (CNNs) require fewer computations for the same accuracy. When observing the data flow in […]

Ver mais

arXiv:2602.00959v1 Announce Type: new Abstract: Large Language Models (LLMs) can be seen as compressed knowledge bases, but it remains unclear what knowledge they truly contain and how far their knowledge boundaries extend. Existing benchmarks are mostly static and provide limited support for systematic knowledge probing. In this paper, we propose an interactive agentic framework to systematically extract and quantify the knowledge of LLMs. Our method includes four adaptive exploration policies to probe knowledge at different granularities. To ensure […]

Ver mais

A question for people working in RL and image generative models (diffusion, flow based etc). There seems to be more emerging work in RL fine tuning techniques for these models (e.g. DDPO, DiffusionNFT, etc). I’m interested to know – is it crazy to try to train these models from scratch with a reward signal only (i.e without any supervision data from a random initialised policy)? And specifically, what techniques could be used to overcome issues with reward sparsity […]

Ver mais

arXiv:2601.08045v1 Announce Type: new Abstract: The widespread adoption of Large Language Models (LLMs) in software development is transforming programming from a solution-generative to a solution-evaluative activity. This shift opens a pathway for new cognitive challenges that amplify existing decision-making biases or create entirely novel ones. One such type of challenge stems from cognitive biases, which are thinking patterns that lead people away from logical reasoning and result in sub-optimal decisions. How do cognitive biases manifest and impact decision-making […]

Ver mais

arXiv:2601.17024v1 Announce Type: new Abstract: The widespread availability of generative artificial intelligence (GenAI) has created a pressing challenge in computer science (CS) education: how to incorporate powerful AI tools into programming coursework without undermining student learning through cognitive offloading. This paper presents an assessment model that permits the use of generative AI for take-home programming assignments while enforcing individual mastery through immediate, assignment-driven written quizzes. To promote authentic learning, these in-class, closed-book assessments are weighted more heavily than […]

Ver mais

arXiv:2601.18947v1 Announce Type: new Abstract: Strong Stability Preserving (SSP) time integration schemes maintain stability of the forward Euler method for any initial value problem. However, only a small subset of Runge-Kutta (RK) methods are SSP, and many efficient high-order time integration schemes do not formally belong to this class. In this work, we introduce a mathematical strategy to analyze the nonlinear stability of RK schemes that may not necessarily belong to the SSP class. With this approach, we […]

Ver mais

The app sec community was happy to see that OWASP is considering making a move in their Top 10 update: “Security Logging and Alerting Failures” from position #10 to position #9, and highlighted in the 2025 release with a new name emphasizing a critical component that was often overlooked—alerting. “Security Logging & Alerting Failures” represents more than a simple reordering of priorities. It signals a shift in how organizations will approach application security in an era of increasingly […]

Ver mais