Meet DigiData: AI that can use your phone Imagine an assistant that taps, swipes, and navigates apps to finish tasks for you. This paper introduces DigiData—a large, diverse, multi-modal dataset built to train mobile control agents to do exactly that. * Richer goals: Instead of scraping random user logs, DigiData
AI
Big win for AI in hidden-information games. Stratego has long resisted top-tier AI—even with training runs costing millions. This new study cracks it. Researchers achieved vastly superhuman Stratego play using self-play reinforcement learning plus a smart test-time search designed for imperfect information. The surprise: it worked with only a
AI
Can we test moderation before it goes live? Researchers built an AI-powered simulator that replays the same online conversation under different moderation rules—like running parallel "what if" worlds—so we can measure how toxic talk changes while keeping everything else equal. * LLM agents act with striking psychological
Robotics
What if robots could learn new tasks just by watching AI-generated videos - and still obey physics? PhysWorld is a new framework that turns text-and-image prompts into physically executable robot skills. * It first generates a task-conditioned video from a single scene image and a command. * Then it reconstructs the scene&
4D3R in a nutshell Turning single‑camera videos of moving scenes into crisp 3D views is hard—especially without known camera poses. 4D3R tackles this with a pose‑free, motion‑aware pipeline. * Two-stage design: first estimates camera and rough geometry using 3D foundation models, then refines with motion cues. * MA‑
Quantum circuits that evolve Researchers propose a gate-based Quantum Genetic Algorithm (QGA) for solving tough global optimization problems. Instead of DNA, each individual is a tiny quantum circuit. When measured, the circuit produces bits that decode into real numbers. The algorithm applies mutation and crossover directly to the gates, in
How good are today’s coding AIs at real developer work—not just toy puzzles? Meet SWE-Compass, a new benchmark that puts large language models (LLMs) through realistic, production-style tasks. What’s new * Broad coverage: 8 task types across 8 development scenarios and 10 programming languages. * Real data: 2,000
Neural networks can be fooled by adversarial images—but here’s a twist: those fake images are even more fragile than the real ones, especially when parts are hidden. Researchers tested nine popular attacks (like FGSM and PGD) on CIFAR-10 and slid a small mask across each image while watching
Most pathology AIs read slides patch by patch, often missing the subtle shapes of individual cells. JWTH (Joint-Weighted Token Hierarchy) changes that by fusing cell-level clues with whole-tissue context in one foundation model. * What’s new: a cell-centric post-tuning step and attention pooling that combine local and global tokens. * Why
As more networks switch to Software-Defined Networking (SDN), keeping them both safe and speedy is a challenge. SmartSecChain-SDN is a new framework that tackles both. It blends three pieces: * AI intrusion detection using Random Forest, XGBoost, CatBoost, and a CNN-BiLSTM to spot attacks in real time with high precision. * Tamper-resistant
AI that writes your architecture docs—from your code Architecture docs often lag behind reality. This research shows a semi-automated way to generate Software Architecture Descriptions (SADs) directly from source code. * Recover the system’s structure with reverse engineering to build a high-level component diagram. * Use a large language model
AI agents are starting to work in teams. That unlocks power—and new ways things can go wrong. TAMAS is a benchmark that stress‑tests multi‑agent LLM systems against adversarial tricks and coordination failures. * 5 realistic scenarios, 300 attack instances across 6 attack types * 211 tools, plus 100 harmless