Search papers, labs, and topics across Lattice.
1
0
2
Language models struggle to predict changes in dynamic environments, achieving only 13.8% accuracy in a rigorous benchmark designed to test causal reasoning and state transitions.
