6 papers
Kimi K3: Open Frontier Intelligence
Kimi Team, Tongtong Bai, Yifan Bai +398
We introduce Kimi K3, a 2.8T parameter Mixture-of-Experts model with 104 billion activated parameters, native vision capabilities, and a 1-million-token context window. Kimi K3 is…
PagedWeight: Efficient MoE LLM Serving with Dynamic Quality-Aware Weight Quantization
Yuchen Yang, Yifan Zhao, Anisha Dasgupta +1
Mixture-of-Experts (MoE) is a popular class of large language models (LLMs), offering high efficiency and accuracy. However, in KV-cache-intensive serving scenarios, MoEs often exh…
AREAL-DTA: Dynamic Tree Attention for Efficient Reinforcement Learning of Large Language Models
Jiarui Zhang, Yuchen Yang, Ran Yan +8
Reinforcement learning (RL)-based post-training for large language models (LLMs) is computationally expensive, as it generates many rollout sequences that frequently share long tok…
D^2SD: Accelerating Speculative Decoding with Dual Diffusion Draft Models
Liyuan Zhang, Jiarui Zhang, Jinwei Yao +6
Speculative decoding accelerates autoregressive large language model inference by drafting multiple tokens and verifying them in a single target-model forward pass. Recent diffusio…
AgentKernelArena: Generalization-Aware Benchmarking of GPU Kernel Optimization Agents
Sharareh Younesian, Wenwen Ouyang, Sina Rafati +11
GPU kernel optimization is increasingly critical for efficient deep learning systems, but writing high-performance kernels still requires substantial low-level expertise. Recent AI…
Nautilus: An Auto-Scheduling Tensor Compiler for Efficient Tiled GPU Kernels
Yifan Zhao, Yuchen Yang, Matei Budiu +1
We present Nautilus, a novel tensor compiler that moves toward fully automated math-to-kernel optimization. Nautilus compiles a high-level algebraic specification of tensor operato…