4 papers · 1 filter
Activation Outliers Matter: Robust Recovery for Quantized Multimodal LLMs
Tanzila Rahman, Mehran Taghian Jazi, Yunke Peng +10
Low-bit quantization offers a promising avenue for reducing the computational and memory demands of Multimodal Large Language Models (MLLMs). Recent hardware support for low-precis…
HiFloat4 Format for End-To-End Reinforcement Learning Post-Training of Large Language Models
Hei Yi Mak, Shadan Golestan, Hoang Le +10
We present, to our knowledge, the first end-to-end FP4 RL post-training, in which both the rollout and training policies, including their forward and backward passes, operate at 4-…
Forget Without Compromise: Nexus Sampling for Streaming KV-Cache Eviction Under Fixed Budgets
Duc Duong, Hoang Anh Duy Le, Jianwen Xie +2
Long-context and agentic LLM workloads push the KV cache past any fixed memory budget, forcing the inference stack to permanently evict tokens at every step of a continuous-inferen…
Scout Before You Attend: Sketch-and-Walk Sparse Attention for Efficient LLM Inference
Hoang Anh Duy Le, Sahil Joshi, Zeyu Yang +2
Self-attention dominates the computational and memory cost of long-context LLM inference across both prefill and decode phases. To address this challenge, we introduce Sketch&Walk…