arXiv:2602.18483v1 Announce Type: new Abstract: Recently, red teaming, with roots in security, has become a key evaluative approach to ensure the safety and reliability of Generative Artificial Intelligence. However, most existing work emphasizes technical benchmarks and attack success rates, leaving the socio-technical practices of how red teaming datasets are defined, created, and evaluated under-examined. Drawing on 22 interviews with practitioners who design and evaluate red teaming datasets, we examine the data practices and standards that underpin this work. […]

Ver mais

arXiv:2602.21228v1 Announce Type: new Abstract: As applications of large language models (LLMs) become increasingly complex, the demand for robust complex instruction following capabilities is growing accordingly. We argue that a thorough understanding of the instruction itself, especially the latent reasoning structure embedded between the lines, is crucial for improving instruction following. Therefore we target complex instructions that involve implicit reasoning, intricate logical relations, and multi-constraint dependencies. We propose ImpRIF, a method to enhance LLMs’ understanding of implicit reasoning […]

Ver mais

Wearable smart inhalers represent a transformative approach to chronic respiratory management in advanced lung cancer, integrating sensor‑based monitoring, real‑time connectivity, and patient‑centric feedback loops. These devices track inhalation technique, dosing frequency, and timing, then transmit encrypted data to cloud‑based platforms for analysis by clinicians and AI‑driven algorithms. For advanced‑stage lung cancer patients, this enables continuous surveillance of bronchospasm, dyspnoea, and medication adherence outside the intensive care setting, thereby reducing uncontrolled exacerbations and unplanned hospitalizations. Smart inhalers also support […]

Ver mais

General aviation fault diagnosis and efficient maintenance are critical to flight safety; however, deploying deep learning models on resource-constrained edge devices poses dual challenges in computational capacity and interpretability. This paper proposes LiteInception–a lightweight interpretable fault diagnosis framework designed for edge deployment. The framework adopts a two-stage cascaded architecture aligned with standard maintenance workflows: Stage 1 performs high-recall fault detection, and Stage 2 conducts fine-grained fault classification on anomalous samples, thereby decoupling optimization objectives and enabling on-demand allocation […]

Ver mais

We study the problem of learning multi-index models (MIMs), where the label depends on the input $boldsymbol{x} in mathbb{R}^d$ only through an unknown $mathsf{s}$-dimensional projection $boldsymbol{W}_*^mathsf{T} boldsymbol{x} in mathbb{R}^mathsf{s}$. Exploiting the equivariance of this problem under the orthogonal group $mathcal{O}_d$, we obtain a sharp harmonic-analytic characterization of the learning complexity for MIMs with spherically symmetric inputs — which refines and generalizes previous Gaussian-specific analyses. Specifically, we derive statistical and computational complexity lower bounds within the Statistical Query (SQ) […]

Ver mais

We study an online learning problem with multiple correct answers, where each instance admits a set of valid labels, and in each round the learner must output a valid label for the queried example. This setting is motivated by language generation tasks, in which a prompt may admit many acceptable completions, but not every completion is acceptable. We study this problem under three feedback models. For each model, we characterize the optimal mistake bound in the realizable setting […]

Ver mais

In this quiz, you’ll test your understanding of AI Coding Agents Guide: A Map of the Four Workflow Types. By working through this quiz, you’ll revisit the four common workflow types for AI coding agents: IDE, terminal, pull request, and cloud. You’ll also get a chance to review how to match the right workflow to the task in front of you. [ Improve Your Python With 🐍 Python Tricks 💌 – Get a short & sweet Python Trick […]

Ver mais

FLUX.2-klein-4B Pure C Implementation On 15th January Black Forest Labs, a lab formed by the creators of the original Stable Diffusion, released black-forest-labs/FLUX.2-klein-4B – an Apache 2.0 licensed 4 billion parameter version of their FLUX.2 family. Salvatore Sanfilippo (antirez) decided to build a pure C and dependency-free implementation to run the model, with assistance from Claude Code and Claude Opus 4.5. Salvatore shared this note on Hacker News: Something that may be interesting for the reader of this […]

Ver mais