5 papers
Libra: Taming Attention Workload Skew in Long-Context LLM Training with Bounded Sequence Pool
Yan Wang, Xiulong Yuan, Kaiming Yang +16
Long-context LLM training suffers from a load-balancing problem that sequence packing does not solve. Packing samples into fixed-token sequences balances memory and linear-cost ope…
SlimQwen: Exploring the Pruning and Distillation in Large MoE Model Pre-training
Shengkun Tang, Zekun Wang, Bo Zheng +7
Structured pruning and knowledge distillation (KD) are typical techniques for compressing large language models, but it remains unclear how they should be applied at pretraining sc…
Accelerating Compound LLM Training Workloads with Maestro
Xiulong Yuan, Hongqing Chen, Jiaxuan Peng +16
Compound LLM training workloads-such as knowledge distillation and multimodal LLM (MLLM) training-are gaining prominence. These typically comprise heterogeneous components differin…
Efficient Long Context Fine-tuning with Chunk Flow
Xiulong Yuan, Hongtao Xu, Wenting Shen +10
Long context fine-tuning of large language models(LLMs) involves training on datasets that are predominantly composed of short sequences and a small proportion of longer sequences.…
BladeDISC++: Memory Optimizations Based On Symbolic Shape
Xiulong Yuan, Xu Yan, Wenting Shen +5
Recent deep learning workloads exhibit dynamic characteristics, leading to the rising adoption of dynamic shape compilers. These compilers can generate efficient kernels for dynami…