9 citations · 11 across the 8 of their papers we have counts for
9 papers
How to Train Your Little Dragon: Teaching a Small Language Model to Construct and Refine GPU Optimization Scripts
Annabelle Sujun Tang, Wei Tang, Songyao Jiang +7
Compiler transformations interact, making GPU kernel optimization a search for combinations that preserve the computation while improving performance. We introduce HOW TO TRAIN YOU…
AccelOpt: A Self-Improving LLM Agentic System for AI Accelerator Kernel Optimization
Genghan Zhang, Shaowei Zhu, Anjiang Wei +6
We present AccelOpt, a self-improving large language model (LLM) agentic system that autonomously optimizes kernels for emerging AI acclerators, eliminating the need for expert-pro…
DCP: Addressing Input Dynamism In Long-Context Training via Dynamic Context Parallelism
Chenyu Jiang, Zhenkun Cai, Ye Tian +3
Context parallelism has emerged as a key technique to support long-context training, a growing trend in generative AI for modern large models. However, existing context parallel me…
PLoRA: Efficient Concurrent LoRA Training for Large Language Models
Minghao Yan, Zhuang Wang, Zhen Jia +2
Low-Rank Adaptation (LoRA) has gained popularity as a fine-tuning approach for Large Language Models (LLMs) due to its low resource requirements and good performance. While numerou…
TTrace: Lightweight Error Checking and Diagnosis for Distributed Training
Haitian Jiang, Shaowei Zhu, Zhen Zhang +5
Distributed training is essential for scaling the training of large neural network models, such as large language models (LLMs), across thousands of GPUs. However, the complexity o…
Marconi: Prefix Caching for the Era of Hybrid LLMs
Rui Pan, Zhuang Wang, Zhen Jia +5
Hybrid models that combine the language modeling capabilities of Attention layers with the efficiency of Recurrent layers (e.g., State Space Models) have gained traction in practic…
DiffusionPipe: Training Large Diffusion Models with Efficient Pipelines
Ye Tian, Zhen Jia, Ziyue Luo +2
Diffusion models have emerged as dominant performers for image generation. To support training large diffusion models, this paper studies pipeline parallel training of diffusion mo…
Lancet: Accelerating Mixture-of-Experts Training via Whole Graph Computation-Communication Overlapping
Chenyu Jiang, Ye Tian, Zhen Jia +3
The Mixture-of-Expert (MoE) technique plays a crucial role in expanding the size of DNN model parameters. However, it faces the challenge of extended all-to-all communication laten…
DynaPipe: Optimizing Multi-task Training through Dynamic Pipelines
Chenyu Jiang, Zhen Jia, Shuai Zheng +2
Multi-task model training has been adopted to enable a single deep neural network model (often a large language model) to handle multiple tasks (e.g., question answering and text s…