Showing cs.AIShow all
2 papers · 1 filter
cs.AI2026
CHARM+: Cross-Hardware Attention with Re-Merge Multistream Mechanism
Xinhang Zhang, Boning Zhang, Chengchun Liu +4
As sequence lengths continue to grow, distributed attention based on sequence parallel (e.g., Ring-Attention) has become a mainstream solution for large language models (LLMs). Nev…
cs.AI2026
CoFlow: Coordinated Few-Step Flow for Offline Multi-Agent Decision Making
Guowei Zou, Haitao Wang, Beiwen Zhang +2
How can generative offline multi-agent reinforcement learning achieve both fast joint trajectory generation and effective cooperation? Multi-agent diffusion models generate joint t…