Search papers, labs, and topics across Lattice.
B-Instruct, it brings the GSM
3
0
4
LOTUS achieves a groundbreaking 2.5x-6.9x reduction in reasoning latency while matching explicit chain-of-thought performance at 3B parameters.
CoT transformers can simulate Word RAM algorithms with poly-logarithmic overhead, revolutionizing our understanding of their computational efficiency.
Causal evaluation reveals that relying on correlational analysis can lead to significant misinterpretations of task learnability in language models.
