4 accepted papers
Mixture-of-Experts architectures have become the standard for efficient LLM scaling, typically employing expert parallelism to distribute experts across devices. However, the absence of explicit load balancing constraints during inference allows adversarial inputs to trigger severe routing concentra…
The frequent retrieval of Key-Value (KV) cache data has emerged as a significant factor contributing to the inefficiency of the inference process in large language models. Previous research has demonstrated that a small subset of critical KV cache tokens largely influences attention outcomes, leadin…
Large Language Models (LLMs) capable of handling extended contexts are in high demand, yet their inference remains challenging due to substantial Key-Value (KV) cache size and high memory bandwidth requirements. Previous research has demonstrated that KV cache exhibits low-rank characteristics withi…
Image text carries essential information to understand the scene and perform reasoning. Text-based visual question answering (text VQA) task focuses on visual questions that require reading text in images. Existing text VQA systems generate an answer by selecting from optical character recognition (…