Search papers, labs, and topics across Lattice.
University of Cambridge
1
0
2
Explicitly regulating a research agent's mental model with feedback loops beats relying on implicit LLM reasoning, leading to significant gains on complex research tasks.
