Simulating Humans to Build Better Robots
A new framework uses detailed human modeling and artificial intelligence to design robots that interact with people more naturally and effectively.
Science
Designing Molecules with the Power of Thought
![Logos integrates the predictive power of specialized chemical models with the reasoning capabilities of large language models, employing a three-stage training pipeline-self-data distillation, supervised fine-tuning, and molecule-focused reinforcement learning-to achieve near-perfect validity scores [latex] \sim99.9\% [/latex] on benchmark datasets like ChEBI-20 and PCdes, and ultimately enabling interactive molecular design.](https://arxiv.org/html/2603.09268v1/x1.png)
A new AI system combines language and chemical reasoning to create promising molecular designs, challenging the need for massive model scales.
Beyond Performance: Understanding How We Experience AI
A new framework proposes that studying the lived experience of interacting with artificial intelligence is crucial to designing systems that truly align with human values and needs.
Learning by Showing: Robots That Ask for the Best Guidance
A new algorithm helps robots actively solicit feedback from humans to quickly learn complex tasks and improve the teaching experience.
Smart Routing for Smarter Robots
A new framework intelligently selects the best robotic policy for any given manipulation task, bypassing the need for extensive training.
Let Robots Explore: A New Path to Complex Manipulation
Researchers have developed a novel approach that allows robots to learn diverse and intricate movements through self-directed exploration, bypassing the need for pre-programmed guidance.
The Human Firewall: Modeling Behavior in a Digital World
Understanding how people react to cyber threats is crucial, and new research suggests these insights can also fortify the defenses of increasingly autonomous AI systems.
Mapping Reality: Robots Embrace Perceptual Ambiguity
A new approach to indoor navigation allows robots to better understand spaces by acknowledging, rather than correcting, the natural perceptual errors inherent in human understanding.
The AI Echo Chamber: What Happens When Bots Socialize?
A new analysis of the first AI-only social network, Moltbook, reveals how artificial intelligence agents communicate when left to their own devices.
Teaching Robots to Check Their Work
A new system uses expert feedback to improve the safety and reliability of robot programs generated by large language models.