arXiv:2602.08210v1 Announce Type: cross Abstract: Heatmap-based solvers have emerged as a promising paradigm for Combinatorial Optimization (CO). However, we argue that the dominant Supervised Learning (SL) training paradigm suffers from a fundamental objective mismatch: minimizing imitation loss (e.g., cross-entropy) does not guarantee solution cost minimization. We dissect this mismatch into two deficiencies: Decoder-Blindness (being oblivious to the non-differentiable decoding process) and Cost-Blindness (prioritizing structural imitation over solution quality). We empirically demonstrate that these intrinsic flaws impose a hard […]

Ver mais

arXiv:2604.14621v2 Announce Type: replace Abstract: Conformal prediction (CP) has attracted broad attention as a simple and flexible framework for uncertainty quantification through prediction sets. In this work, we study how to deploy CP under differential privacy (DP) in a statistically efficient manner. We first introduce differential CP, a non-splitting conformal procedure that avoids the efficiency loss caused by data splitting and serves as a bridge between oracle CP and private conformal inference. By exploiting the stability properties of […]

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:2602.12430v1 Announce Type: new Abstract: The transition from monolithic language models to modular, skill-equipped agents marks a defining shift in how large language models (LLMs) are deployed in practice. Rather than encoding all procedural knowledge within model weights, agent skills — composable packages of instructions, code, and resources that agents load on demand — enable dynamic capability extension without retraining. It is formalized in a paradigm of progressive disclosure, portable skill definitions, and integration with the Model Context […]

Ver mais

If you’ve ever built a data pipeline for analytics or business intelligence, you know the basics — ingest, transform, store. But regulated industries are a different game entirely. A missed record in a BI pipeline means a slightly off dashboard. A missed record in a compliance pipeline means a regulatory fine, a failed audit, or worse. Over the past couple of years, I’ve been designing and running ETL pipelines that process over 500,000 compliance data files every month […]

Ver mais

arXiv:2602.18459v1 Announce Type: new Abstract: LLMs act in the social world by drawing upon shared cultural patterns to make social situations understandable and actionable. Because identity is often part of the inferential substrate of competent judgment, ethical alignment requires regulating when and how systems invoke identity. Yet the dominant governance regime for identity-related harm remains bias mitigation, which treats identity primarily as a source of measurable disparities or harmful associations to be detected and suppressed. This leaves underspecified […]

Ver mais