Search papers, labs, and topics across Lattice.
The Hong Kong Polytechnic University
3
0
4
Co-evolving world models with agent policies leads to a 16.75% boost in performance, revolutionizing how language agents navigate complex tasks.
Active belief intervention can drastically improve embodied agents' decision-making by overcoming the pitfalls of belief inertia.
Achieve 3x higher success in long-horizon robotic manipulation by explicitly separating high-level planning from low-level control, enabling memory-aware reasoning and adaptive replanning.
