Join us as we discuss Accurate KV Cache Quantization with Outlier Tokens Tracing, a deep dive into improving the efficiency of LLM inference. The authors enhance KV Cache quantization, a technique for reducing memory and compute costs during inference, by introducing a method to identify and exclude outlier tokens that hurt quantization accuracy, striking a better balance between efficiency and performance. Paper: https://arxiv.org/abs/2505.10938 Slides: https://bit.ly/45wolpr Join us for Ar...
All content for Deep Papers is the property of Arize AI and is served directly from their servers
with no modification, redirects, or rehosting. The podcast is not affiliated with or endorsed by Podjoint in any way.
Join us as we discuss Accurate KV Cache Quantization with Outlier Tokens Tracing, a deep dive into improving the efficiency of LLM inference. The authors enhance KV Cache quantization, a technique for reducing memory and compute costs during inference, by introducing a method to identify and exclude outlier tokens that hurt quantization accuracy, striking a better balance between efficiency and performance. Paper: https://arxiv.org/abs/2505.10938 Slides: https://bit.ly/45wolpr Join us for Ar...
Sleep-time Compute: Beyond Inference Scaling at Test-time
Deep Papers
30 minutes
1 month ago
Sleep-time Compute: Beyond Inference Scaling at Test-time
What if your LLM could think ahead—preparing answers before questions are even asked? In this week's paper read, we dive into a groundbreaking new paper from researchers at Letta, introducing sleep-time compute: a novel technique that lets models do their heavy lifting offline, well before the user query arrives. By predicting likely questions and precomputing key reasoning steps, sleep-time compute dramatically reduces test-time latency and cost—without sacrificing performance. We explore n...
Deep Papers
Join us as we discuss Accurate KV Cache Quantization with Outlier Tokens Tracing, a deep dive into improving the efficiency of LLM inference. The authors enhance KV Cache quantization, a technique for reducing memory and compute costs during inference, by introducing a method to identify and exclude outlier tokens that hurt quantization accuracy, striking a better balance between efficiency and performance. Paper: https://arxiv.org/abs/2505.10938 Slides: https://bit.ly/45wolpr Join us for Ar...
