6 accepted papers
The rapid scaling of large language models (LLMs) has made distributed inference indispensable, yet end-to-end latency is increasingly dominated by communication, forming a critical bandwidth wall that fundamentally limits the practical gains of existing quantization techniques. Existing approaches …
The scaling of large language models (LLMs) is currently bottlenecked by the rigidity of distributed programming. While high-performance libraries like CuBLAS and NCCL provide optimized primitives, they lack the flexibility required for rapidly evolving model architectures. Conversely, existing tens…
The quadratic complexity of softmax attention poses a major bottleneck for long-context modeling, motivating a surge of linear attention variants with linear complexity. Unlike softmax attention, which benefits from optimized kernels, linear attention lacks general-purpose, hardware-efficient suppor…
Reinforcement learning (RL) is crucial for aligning Vision-Language Models (VLMs), but its practical application is hampered by significant system-level bottlenecks. The typical RL pipeline, encompassing data loading, inference-based rollouts, and model updates, suffers from severe inefficiencies wh…
Mixture-of-Experts (MoE) models face deployment challenges due to their large parameter counts and computational demands. We explore quantization for MoE models and highlight two key insights: 1) linear blocks exhibit varying quantization sensitivity, and 2) divergent expert activation frequencies c…
Large language models (LLMs) have demonstrated remarkable performance, and organizations are racing to serve LLMs of varying sizes as endpoints for use-cases like chat, programming and search. However, efficiently serving multiple LLMs poses significant challenges for existing approaches due to vary…