Search papers, labs, and topics across Lattice.
2
0
5
2
ReTopK achieves a 3.07x speedup in attention computation with only a 0.50% increase in perplexity, revolutionizing long-context processing efficiency.
By normalizing rewards across groups of sampled communication graphs, Graph-GRPO stabilizes multi-agent topology learning and uncovers critical communication pathways obscured by noisy, absolute rewards.
