How Children and Adults Learn Causality via Empowerment
Explore how children and adults use empowerment, controllability, and variability to build causal models and enhance learning.
Saber: Fast, High-Quality Sampling for Diffusion Language Models
Discover Saber, a training-free sampling method boosting speed and accuracy in diffusion language models for efficient code generation.
Ensuring Safety and Security in Agentic AI Systems
Discover a unified framework formalizing safety, security, and functionality in agentic AI systems for reliable multi-agent coordination.
ProRe: Proactive Reward System for GUI Agents
Discover ProRe, a proactive reward system enhancing GUI agent training via reasoner-actor collaboration for improved accuracy and performance.
MAS-Bench: Benchmark for Hybrid Mobile GUI Agents
Discover MAS-Bench, a unified benchmark for evaluating shortcut-augmented hybrid mobile GUI agents with 139 tasks and 9 metrics.