6 papers · 1 filter
Memory Caching: RNNs with Growing Memory
Ali Behrouz, Zeman Li, Yuan Deng +3
Transformers have been established as the de-facto backbones for most recent advances in sequence modeling, mainly due to their growing memory capacity that scales with the context…
TNT: Improving Chunkwise Training for Test-Time Memorization
Zeman Li, Ali Behrouz, Yuan Deng +5
Recurrent neural networks (RNNs) with deep test-time memorization modules, such as Titans and TTT, represent a promising, linearly-scaling paradigm distinct from Transformers. Whil…
Synthetic Text Generation for Training Large Language Models via Gradient Matching
Dang Nguyen, Zeman Li, Mohammadhossein Bateni +3
Synthetic data has the potential to improve the performance, training efficiency, and privacy of real training examples. Nevertheless, existing approaches for synthetic text genera…
PiKE: Adaptive Data Mixing for Large-Scale Multi-Task Learning Under Low Gradient Conflicts
Zeman Li, Yuan Deng, Peilin Zhong +2
Modern foundation models are trained on diverse datasets to enhance generalization across tasks and domains A central challenge in this process is determining how to effectively mi…
Addax: Utilizing Zeroth-Order Gradients to Improve Memory Efficiency and Performance of SGD for Fine-Tuning Language Models
Zeman Li, Xinwei Zhang, Peilin Zhong +3
Fine-tuning language models (LMs) with the Adam optimizer often demands excessive memory, limiting accessibility. The "in-place" version of Stochastic Gradient Descent (IP-SGD) and…
Optimal Differentially Private Model Training with Public Data
Andrew Lowy, Zeman Li, Tianjian Huang +1
Differential privacy (DP) ensures that training a machine learning model does not leak private data. In practice, we may have access to auxiliary public data that is free of privac…