arXiv:2601.20895v1 Announce Type: new Abstract: Research aimed at scaling up neuroscience inspired learning algorithms for neural networks is accelerating. Recently, a key research area has been the study of energy-based learning algorithms such as predictive coding, due to their versatility and mathematical grounding. However, the applicability of such methods is held back by the large computational requirements caused by their iterative nature. In this work, we address this problem by showing that the choice of initialization of the […]

Ver mais

arXiv:2603.25943v1 Announce Type: new Abstract: Enormous fluid antenna systems (E-FAS) have recently emerged as a surface-wave (SW)-enabled architecture that can induce controllable large-scale channel gains through guided electromagnetic routing. This paper develops a secrecy analysis framework for E-FAS-assisted downlink transmission with practical pilot-based channel estimation. We consider a multiple-input single-output (MISO) wiretap setting in which the base station (BS) performs minimum mean-square-error (MMSE) channel estimation and adopts maximum-ratio transmission (MRT) with artificial noise (AN). To capture the leakage […]

Ver mais

What if you could hold a hedge fund position and still trade its liquidity on a public blockchain at 3 a.m. on a Sunday? That is the question Tok-Edge, a London-based digital asset firm, is putting in front of institutional allocators with the launch of what it calls the Redemption Token, confirmed on April 7, 2026 at a $15 million company valuation. The firm raised roughly $1.5 million in its pre-launch round, led by Marcus Meijer, founder of […]

Ver mais

The sample complexity of multi-group learning is shown to improve in the group-realizable setting over the agnostic setting, even when the family of groups is infinite so long as it has finite VC dimension. The improved sample complexity is obtained by empirical risk minimization over the class of group-realizable concepts, which itself could have infinite VC dimension. Implementing this approach is also shown to be computationally intractable, and an alternative approach is suggested based on improper learning.

Ver mais

arXiv:2603.24631v1 Announce Type: new Abstract: Code agents can autonomously resolve GitHub issues, yet when they fail, current evaluation provides no visibility into where or why. Metrics such as Pass@1 collapse an entire execution into a single binary outcome, making it difficult to identify where and why the agent went wrong. To address this limitation, we introduce TRAJEVAL, a diagnostic framework that decomposes agent trajectories into three interpretable stages: search (file localization), read (function comprehension), and edit (modification targeting). […]

Ver mais

arXiv:2602.16805v1 Announce Type: new Abstract: Code evolution is a family of techniques that rely on large language models to search through possible computer programs by evolving or mutating existing code. Many proposed code evolution pipelines show impressive performance but are often not compared to simpler baselines. We test how well two simple baselines do over three domains: finding better mathematical bounds, designing agentic scaffolds, and machine learning competitions. We find that simple baselines match or exceed much more […]

Ver mais

arXiv:2603.22332v1 Announce Type: new Abstract: Data imputation is a cornerstone technique for handling missing values in real-world datasets, which are often plagued by missingness. Despite recent progress, prior studies on Large Language Models-based imputation remain limited by scalability challenges, restricted cross-model comparisons, and evaluations conducted on small or domain-specific datasets. Furthermore, heterogeneous experimental protocols and inconsistent treatment of missingness mechanisms (MCAR, MAR, and MNAR) hinder systematic benchmarking across methods. This work investigates the robustness of Large Language Models […]

Ver mais