Search papers, labs, and topics across Lattice.
4
0
7
0
FP4 RL for MoE LLMs can now match BF16 rollout quality with 5.4x speedup by directly aligning train-rollout quantization.
Achieving superior performance with one-third the resources, Qwen3.8-Flash-Next redefines efficiency in large-scale language models.
Native computer use can be achieved at scale, enabling agents to outperform leading systems while significantly enhancing security against adversarial attacks.
MTP acceptance rates can be dramatically improved by addressing entropy fluctuations, leading to up to 1.8x faster RL training.
