Self-supervised learning has emerged as a major technique for the task of learning from unlabeled data, where the current methods mostly revolve around alignment of representations and input recon struction. Although such approaches have demonstrated excellent performance in practice, their scope remains mostly confined to learning from observed data and does not provide much help in terms of a learning structure that is predictive of the data distribution. In this paper, we study some of the recent developments […]

Ver mais

arXiv:2508.07473v3 Announce Type: replace-cross Abstract: In Online Convex Optimization (OCO), when the stochastic gradient has a finite variance, many algorithms provably work and guarantee a sublinear regret. However, limited results are known if the gradient estimate has a heavy tail, i.e., the stochastic gradient only admits a finite $mathsf{p}$-th central moment for some $mathsf{p}inleft(1,2right]$. Motivated by it, this work examines different old algorithms for OCO (e.g., Online Gradient Descent) in the more challenging heavy-tailed setting. Under the standard […]

Ver mais

arXiv:2602.12322v1 Announce Type: new Abstract: Vision-Language-Action (VLA) models convert high-level language instructions into concrete, executable actions, a task that is especially challenging in open-world environments. We present Visual Foresight Planning (ForeAct), a general and efficient planner that guides a VLA step-by-step using imagined future observations and subtask descriptions. With an imagined future observation, the VLA can focus on visuo-motor inference rather than high-level semantic reasoning, leading to improved accuracy and generalization. Our planner comprises a highly efficient foresight […]

Ver mais

arXiv:2601.20893v1 Announce Type: new Abstract: Background: Systematic literature reviews (SLRs) have become prevalent in software engineering research. Several researchers may conduct SLRs on similar topics without a prospective register for SLR protocols. However, even ignoring these unavoidable duplications of effort in the simultaneous conduct of SLRs, the proliferation of overlapping and often repetitive SLRs indicates that researchers are not extensively checking for existing SLRs on a topic. Given how effort-intensive it is to design, conduct, and report an […]

Ver mais

arXiv:2604.03858v2 Announce Type: replace-cross Abstract: Training Data Attribution (TDA) seeks to trace model predictions back to influential training examples, enhancing interpretability and safety. We formulate TDA as a Bayesian information-theoretic problem: subsets are scored by the information loss they induce – the entropy increase at a query when removed. This criterion credits examples for resolving predictive uncertainty rather than label noise. To scale to modern networks, we approximate information loss using a Gaussian Process surrogate built from tangent […]

Ver mais

arXiv:2601.09853v1 Announce Type: new Abstract: Real-world health questions from patients often unintentionally embed false assumptions or premises. In such cases, safe medical communication typically involves redirection: addressing the implicit misconception and then responding to the underlying patient context, rather than the original question. While large language models (LLMs) are increasingly being used by lay users for medical advice, they have not yet been tested for this crucial competency. Therefore, in this work, we investigate how LLMs react to […]

Ver mais

arXiv:2601.00223v1 Announce Type: new Abstract: We introduce JP-TL-Bench, a lightweight, open benchmark designed to guide the iterative development of Japanese-English translation systems. In this context, the challenge is often “which of these two good translations is better?” rather than “is this translation acceptable?” This distinction matters for Japanese-English, where subtle choices in politeness, implicature, ellipsis, and register strongly affect perceived naturalness. JP-TL-Bench uses a protocol built to make LLM judging both reliable and affordable: it evaluates a candidate […]

Ver mais

Key Highlights: The internet has always struggled to figure out who’s real and who’s not, but now that AI-generated profiles are everywhere, the issue’s gotten worse. There are convincing synthetic identities that act just like real people everywhere. Sam Altman’s World project sees itself as the fix, aiming to build a universal system to separate humans from machines. The newest step is to plug World ID into Tinder, one of the biggest dating platforms out there. Tinder’s betting […]

Ver mais