6 papers
Bridging Draft Policy Misalignment: Group Tree Optimization for Speculative Decoding
Shijing Hu, Jingyang Li, Zhihui Lu +1
Speculative decoding accelerates large language model (LLM) inference by letting a lightweight draft model propose multiple tokens that the target model verifies in parallel. Yet e…
GRIFFIN: Effective Token Alignment for Faster Speculative Decoding
Shijing Hu, Jingyang Li, Xingyu Xie +3
Speculative decoding accelerates inference in large language models (LLMs) by generating multiple draft tokens simultaneously. However, existing methods often struggle with token m…
Memory-Efficient 4-bit Preconditioned Stochastic Optimization
Jingyang Li, Kuangyu Ding, Kim-Chuan Toh +1
Preconditioned stochastic optimization algorithms, exemplified by Shampoo, outperform first-order optimizers by offering theoretical convergence benefits and practical gains in lar…
Towards Understanding Why FixMatch Generalizes Better Than Supervised Learning
Jingyang Li, Jiachun Pan, Vincent Y. F. Tan +2
Semi-supervised learning (SSL), exemplified by FixMatch (Sohn et al., 2020), has shown significant generalization advantages over supervised learning (SL), particularly in the cont…
Towards Understanding Why Data Augmentation Improves Generalization
Jingyang Li, Jiachun Pan, Kim-Chuan Toh +1
Data augmentation is a cornerstone technique in deep learning, widely used to improve model generalization. Traditional methods like random cropping and color jittering, as well as…
LoCo: Low-Bit Communication Adaptor for Large-scale Model Training
Xingyu Xie, Zhijie Lin, Kim-Chuan Toh +1
To efficiently train large-scale models, low-bit gradient communication compresses full-precision gradients on local GPU nodes into low-precision ones for higher gradient synchroni…