The Deep Learning Recipe for LLM Applications
Abstract: Large Language Models (LLMs) have revolutionized AI research and enabled exciting applications. To build a complex LLM application, such as an LLM agent, most existing research relies on insights from other domains or heuristics to manually build the application. However, this approach often requires heavy hand-engineering and fails to fully optimize for the downstream task of interest. Inspired by the tremendous success of deep learning, we proposed to construct LLM applications in a modular manner, similar to building a deep neural network. Our key insight is to make analogies between LLM building blocks, such as retrievals, memories, and prompting strategies, and the successful deep learning modules, such as MLPs, attention, and recurrent modules. We further design forward inference and feedback mechanisms for LLMs, where prompts in LLMs are considered as the weights in deep models, and the prompt optimization from feedback is analogous to the back-propagation algorithm. We additionally leverage a search algorithm to search for the best configuration of LLM applications, similar to the neural architecture search (NAS) in deep learning research. Comprehensive experimental results demonstrate that the proposed deep learning recipe for LLM applications is highly effective, in particular: (1) Organizing LLM modules into deep-learning-style architectures yields noticeable performance gain; (2) Automatic prompt optimization, equivalent to backpropagation, is efficient in incorporating feedback from the task of interest and achieves at least 5% performance improvement; (3) NAS equivalent algorithm works well for further optimizing the LLM application architecture with 11% performance gain compared with randomly designed architectures. Overall, our research demonstrates the exciting opportunity of transferring the success of deep learning to building LLM applications.
Submission Length: Regular submission (no more than 12 pages of main content)
Assigned Action Editor: ~Rui_Zhang7
Submission Number: 3731
Loading