Search papers, labs, and topics across Lattice.
6
0
9
Weak supervisors no longer bottleneck stronger models when their guidance is used solely to accelerate verifier-aligned policy gradients rather than dictate optimization targets.
Static memory models are holding back performance鈥擯roteus reveals that incrementally activating memory can drastically enhance context retention and reduce interference.
Self-distillation may boost accuracy but comes at the hidden cost of significantly reduced output diversity, risking performance in diverse scenarios.
Iteratively training on a self-selected dataset can dramatically enhance vision-language model performance without the need for extra data or pre-training.
Allocating more capacity to earlier layers in language models can significantly enhance performance, challenging the long-held uniform layer design paradigm.
Looped LLMs don't just perform better reasoning, they also internally mirror the distinct inference stages of standard feedforward models, repeating them cyclically.
