- agents
- multimodal-ai
- ai-salon
- research
•
•
•
-
Perception, Evidence & Efficiency for AI in the Physical World
Three pillars for Ambient Physical AI: motion prediction, verifiable trust, and streaming efficiency—enabling human augmentation across AI glasses and assistive robotics.
-
My talks at CVPR 2026 workshops
This posts accompanies my talks at CVPR workshops. We explore ideas around ambient intelligence, visual task assistants and efficiently scaling transformers.
-
Agentic Ambient Intelligence: Bringing AI into the Physical World
We explore the future of Agentic Ambient Intelligence and discover how advancements in perception, reasoning, and motion-guided control are bridging the gap between digital AI capabilities and real-world physical assistance
-
AI Agents: From Language to Multimodal Reasoning
An outline of our recent journey on AI Agents
-
Level up your Agents: Teaching Vision-Language Models to Play by the Rules
We explore how Vision-Language Models can be improved for interactive decision-making by using our new reinforcement learning technique called Advantage-Filtered Supervised Fine-Tuning.