AI
Arabic AI users are hitting a hidden speed bump LLM-powered agents are everywhere—but most tests are English-first. Kubrak et al. present the first benchmark to evaluate tool-calling and agentic skills when users prompt in Arabic. * What they built: a standardized way to measure functional accuracy and robustness in Arabic
AI
AI assistants that remember our past chats can feel personalized, but there is a catch. Using all history can anchor them to old habits; using none makes them forget who you are. SteeM (Steerable Memory Agent) introduces a simple idea: a controllable memory dial. * Fresh-start mode: prioritize new ideas and
AI
Many headlines say AI chatbots "leak" your personal info. This study asks: is that true memorization, or are prompts giving the answer away? The authors find many "leaks" are driven by surface cues in the prompt. They introduce Cue-Resistant Memorization (CRM): test models under low-cue conditions
AI
Turning jailbreakers into the trapped Jailbreak prompts can trick chatbots into breaking rules. HoneyTrap flips the script: a deceptive, multi-agent defense that engages attackers, wastes their time, and keeps helpful answers flowing for regular users. * Threat Interceptor: spots risky prompts early. * Misdirection Controller: steers attackers into harmless honeypots. * Forensic Tracker:
NLP
Ever wish reviews told you exactly how people feel about the camera, battery, or price? That's aspect-based sentiment analysis (ABSA). Arctic-ABSA is a new family of reasoning-infused models built for real-world ABSA. Trained on a large blend of public and carefully generated synthetic data—about 20× the classic
AI
Can we reliably spot AI-written text? This study says yes—when we evaluate detectors the right way. Researchers built a unified, fair benchmark for AI-text detection by combining HC3 and DAIGT v2 and using a topic-based split to prevent “memorizing” subjects. That makes models prove they generalize to new domains.
AI
AI assistants that click, type, and navigate apps need practice worlds — but real, trainable web environments are scarce. InfiniteWeb changes that by automatically generating realistic, multi-page websites at scale so GUI agents can learn real tasks, not toy demos. How it works * Unified specs: A single blueprint ties pages, flows,
AI
Ever wished screen-reading AI would stop scanning every pixel and just tap what you asked for? FocusUI is a new method that helps vision-language models focus on what matters in app and web UIs—speeding them up without losing spatial precision. How it works: * Selects only instruction-relevant patches, down-weighting big
AI
Multi-agent AI systems can solve complex tasks, but over long runs they may slowly veer off course — a phenomenon this paper calls agent drift. As interactions pile up, behavior degrades, decisions slip, and agents stop aligning with each other. * Semantic drift: answers stray from the original intent. * Coordination drift: consensus
NLP
Arctic-ABSA brings aspect-based sentiment analysis closer to real life. Instead of only “positive/negative/neutral,” it adds nuance, multilingual reach, and reasoning for better business decisions. * Five classes: positive, negative, neutral, mixed, and unknown. * Finds aspect-level opinions and the overall tone in one pass. * Multilingual: one model holds 87–91%
AI
LLMs can learn to reason—without task verifiers Many real-world problems don’t have automatic checkers to grade answers, even though we have lots of expert solutions. RARO (Relativistic Adversarial Reasoning Optimization) shows how to train reasoning skills from those examples alone. How it works: * A policy (the model) tries
DrugRepurposing
Meet ChatDRex Finding new uses for approved drugs can save years and millions. But making those predictions usually takes teams of specialists and a tangle of tools. ChatDRex changes that. It’s a conversation-based, no‑code, multi‑agent system that lets clinicians and researchers ask complex bioinformatics questions in plain