399 citations · 4k across the 100 of their papers we have counts for
9 papers · 1 filter
Bridging Discrete and Backpropagation: Straight-Through and Beyond
Liyuan Liu, Chengyu Dong, Xiaodong Liu +2
Backpropagation, the cornerstone of deep learning, is limited to computing gradients for continuous variables. This limitation poses challenges for problems involving discrete late…
Explaining Patterns in Data with Language Models via Interpretable Autoprompting
Chandan Singh, John X. Morris, Jyoti Aneja +2
Large language models (LLMs) have displayed an impressive ability to harness natural language to perform complex tasks. In this work, we explore whether we can leverage this learne…
Learning Math Reasoning from Self-Sampled Correct and Partially-Correct Solutions
Ansong Ni, Jeevana Priya Inala, Chenglong Wang +4
Pretrained language models have shown superior performance on many natural language processing tasks, yet they still struggle at multi-step formal reasoning tasks like grade school…
Sparsely Activated Mixture-of-Experts are Robust Multi-Task Learners
Shashank Gupta, Subhabrata Mukherjee, Krishan Subudhi +4
Traditional multi-task learning (MTL) methods use dense networks that use the same set of shared weights across several different tasks. This often creates interference where two o…
METRO: Efficient Denoising Pretraining of Large Scale Autoencoding Language Models with Model Generated Signals
Payal Bajaj, Chenyan Xiong, Guolin Ke +7
We present an efficient method of pretraining large-scale autoencoding language models using training signals generated by an auxiliary model. Originated in ELECTRA, this training…
Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer
Greg Yang, Edward J. Hu, Igor Babuschkin +7
Hyperparameter (HP) tuning in deep learning is an expensive process, prohibitively so for neural networks (NNs) with billions of parameters. We show that, in the recently discovere…
Compositional Processing Emerges in Neural Networks Solving Math Problems
Jacob Russin, Roland Fernandez, Hamid Palangi +4
A longstanding question in cognitive science concerns the learning mechanisms underlying compositionality in human cognition. Humans can infer the structured relationships (e.g., g…
Adversarial Regularization as Stackelberg Game: An Unrolled Optimization Approach
Simiao Zuo, Chen Liang, Haoming Jiang +5
Adversarial regularization has been shown to improve the generalization performance of deep learning models in various natural language processing tasks. Existing works usually for…
Understanding the Difficulty of Training Transformers
Liyuan Liu, Xiaodong Liu, Jianfeng Gao +2
Transformers have proved effective in many NLP tasks. However, their training requires non-trivial efforts regarding designing cutting-edge optimizers and learning rate schedulers…