arXiv:2606.13684v1 Announce Type: new Abstract: Automatic Bloom’s taxonomy classification of assessment questions can substantially reduce instructor workload, but labeling is subjective and teacher-dependent. Prior machine learning (ML) and deep learning (DL) approaches reported strong within-dataset results, yet were rarely evaluated in cross-dataset settings, leaving real-world generalizability unclear; meanwhile, LLM effectiveness for Bloom question classification has not been systematically studied. We evaluated the cross-dataset generalization of existing ML/DL methods and assessed LLMs with multiple prompting strategies on five datasets; […]

Ver mais

Introduction When I started exploring Retrieval-Augmented Generation (RAG) systems for incident analysis, I realized that jumping straight into paid APIs like Claude or OpenAI wasn’t practical for learning and experimentation. Instead, I wanted to build something completely local, free to run, and powerful enough to handle real production scenarios. This article documents my journey building a fully functional RAG system that analyzes production incidents by learning from past issues — without spending a dime on API calls. Everything runs on […]

Ver mais

Anthropic published a lawyer-related plugin on 30 January, and within four to five days, software stocks had lost billions of dollars. The pace of the collapse left everyone shocked, as the whole industry had been operating on a fundamental misperception of how the AI economy works. Why Everyone Thought They Were Safe The narrative that dominated the technological field held that AI companies would build baseline models, which would then serve as a foundation for software companies to […]

Ver mais

In the high-stakes bazaar of the dark web, there is a clear hierarchy of value. A stolen credit card is a mere commodity, often trading for the price of a cup of coffee. A Social Security number is a versatile tool. But a comprehensive medical record? That is a masterpiece. It contains a permanent, unchangeable blueprint of a human life — history, genetic markers, and biometric vulnerabilities. Unlike a compromised bank account, which can be frozen and reset, […]

Ver mais

arXiv:2603.26688v1 Announce Type: new Abstract: Peer-to-peer energy trading among electric vehicles (EVs) has been increasingly studied as a promising solution for improving supply-side resilience under growing charging demand and constrained charging infrastructure. While prior studies on EV-EV energy trading and related EV research have largely focused on transaction management or isolated mobility prediction tasks, the problem of identifying which charging nodes are more suitable for EV-EV trading in journey contexts remains open. We address this gap by formulating […]

Ver mais

arXiv:2601.01147v2 Announce Type: replace Abstract: Conformal Test Martingales (CTMs) are a standard method within the Conformal Prediction framework for testing the crucial assumption of data exchangeability by monitoring deviations from uniformity in the p-value sequence. Although exchangeability implies uniform p-values, the converse does not hold. This raises the question of whether a significant break in exchangeability can occur, such that the p-values remain uniform, rendering CTMs blind. We answer this affirmatively, demonstrating the phenomenon of emph{conformal blindness}. Through […]

Ver mais

A large number of heuristics have been proposed to optimize the reinforcement fine-tuning of LLMs. However, inconsistent claims are made from time to time, making this area elusive. Reflecting on this situation, two fundamental questions still lack a clear understanding: 1) what is the role of each optimizing choice? 2) which ones are the bottlenecks? This paper aims to shed light on them, and it faces the challenge of several entangled confounding factors in the fine-tuning process. To […]

Ver mais

arXiv:2603.22379v1 Announce Type: new Abstract: Adapters are often selected and deployed based on nominal labels (e.g., instruction-tuned), which implicitly suggest what capability improves after adaptation. We test whether nominal training objectives reliably align with realized cross-task capability gains by evaluating the same LoRA adapter across tasks. Our strongest evidence is tied to strict, automatically verifiable instruction following as measured by IFEval: across multiple seeds, base models, and LoRA settings, nominal labels recurrently but not universally fail to predict […]

Ver mais