Search papers, labs, and topics across Lattice.
University of Edinburgh
4
0
5
Text world models can transform LLM-based agents from reactive responders to proactive planners, fundamentally changing how they interact with complex environments.
A single round of targeted feedback can boost DRA performance by up to 15 points, but subsequent revisions may undo prior gains.
LLM agents can get 18% better at tasks by co-evolving their skills and tools, instead of learning them separately.
LLMs can reason more effectively and efficiently by internalizing tool knowledge, eliminating the need for external documentation and reducing inference costs.
