Blog(2)

• LookaheadKV: Fast and Accurate KV Cache Eviction by Glimpsing into the Future without Generation

  The deployment of Large Language Models (LLMs) for long-context applications is fundamentally constrained by the memory footprint of the Key-Value (KV) cache. As the sequence length increases, the memory required to store KV states grows linearly, leading to severe GPU memory bottlenecks and degraded inference throughput.

• The Resolution Hypothesis: Discovering how Time Scale Maps to Health Outcomes

  Wearable photoplethysmography (PPG) has quietly become one of the most information-dense and ubiquitous sensing modalities in modern health and wellness.

Research Areas(0)

Publications(0)

News(0)

Others(0)