5 papers
Linear Spatial World Models Emerge in Large Language Models
Matthieu Tehenan, Christian Bolivar Moya, Tenghai Long +1
Large language models (LLMs) have demonstrated emergent abilities across diverse tasks, raising the question of whether they acquire internal world models. In this work, we investi…
Adversarial Autoencoders in Operator Learning
Dustin Enyeart, Guang Lin
DeepONets and Koopman autoencoders are two prevalent neural operator architectures. These architectures are autoencoders. An adversarial addition to an autoencoder have improved pe…
Some Best Practices in Operator Learning
Dustin Enyeart, Guang Lin
Hyperparameters searches are computationally expensive. This paper studies some general choices of hyperparameters and training methods specifically for operator learning. It consi…
Loss Terms and Operator Forms of Koopman Autoencoders
Dustin Enyeart, Guang Lin
Koopman autoencoders are a prevalent architecture in operator learning. But, the loss functions and the form of the operator vary significantly in the literature. This paper presen…
LLM Reasoning Engine: Specialized Training for Enhanced Mathematical Reasoning
Shuguang Chen, Guang Lin
Large Language Models (LLMs) have shown remarkable performance in various natural language processing tasks but face challenges in mathematical reasoning, where complex problem-sol…