RedKnot: Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttention
Yang Liu, ZhaoKai Luo, HuaYi Jin, ZhiYong Wang, RuoZhou He, BoYu Wang et al. · arXiv: 2606.06256v1 · Summary page
Your journal club presentation, ready in a minute. About ten slides with the figures of the paper when its license allows it, what to say on each one, the strengths and limits of the paper, and questions to start the discussion. Present it full screen or save it as PDF.