Scaling Cross-Embodied Learning: One Policy for Manipulation, Navigation, Locomotion and Aviation
Paper
•
2408.11812
•
Published
•
4
Embodied AI
Embodiment enables interaction of model with environment. Key is to anticipate what change could've come with its current action.
-
-
WebArena: A Realistic Web Environment for Building Autonomous Agents
Paper • 2307.13854 • Published • 23 -
Agent Workflow Memory
Paper • 2409.07429 • Published • 25 -
Windows Agent Arena: Evaluating Multi-Modal OS Agents at Scale
Paper • 2409.08264 • Published • 39 -
VisualAgentBench: Towards Large Multimodal Models as Visual Foundation Agents
Paper • 2408.06327 • Published • 13 -
Tree Search for Language Model Agents
Paper • 2407.01476 • Published -
DigiRL: Training In-The-Wild Device-Control Agents with Autonomous Reinforcement Learning
Paper • 2406.11896 • Published • 18 -
Fine-Tuning Large Vision-Language Models as Decision-Making Agents via Reinforcement Learning
Paper • 2405.10292 • Published • 1 -
Cognitive Kernel: An Open-source Agent System towards Generalist Autopilots
Paper • 2409.10277 • Published • 1 -
UI-JEPA: Towards Active Perception of User Intent through Onscreen User Activity
Paper • 2409.04081 • Published • 3 -
Achieving Human Level Competitive Robot Table Tennis
Paper • 2408.03906 • Published • 26 -
Multimodal Fusion with LLMs for Engagement Prediction in Natural Conversation
Paper • 2409.09135 • Published • 1