Search papers, labs, and topics across Lattice.
2
0
4
LLMs struggle with compositional reasoning, showing sharp performance drops in executing order-sensitive data refinement tasks.
Directional inconsistency can destabilize LLM training, but geoalign curates rollouts to improve performance and stability, outperforming several established methods.
