Supplement Generation Training for Enhancing Agentic Task Performance
This addresses the problem of high computational costs and rapid obsolescence for developers and organizations using LLM-powered agents, though it is an incremental improvement over existing methods.
The paper tackles the inefficiency of training large foundation models for agentic tasks by proposing Supplement Generation Training (SGT), where a smaller LLM generates supplemental text to enhance a larger LLM's performance without retraining it, resulting in more flexible and cost-effective deployment.
Training large foundation models for agentic tasks is increasingly impractical due to the high computational costs, long iteration cycles, and rapid obsolescence as new models are continuously released. Instead of post-training massive models for every new task or domain, we propose Supplement Generation Training (SGT), a more efficient and sustainable strategy. SGT trains a smaller LLM to generate useful supplemental text that, when appended to the original input, helps the larger LLM solve the task more effectively. These lightweight models can dynamically adapt supplements to task requirements, improving performance without modifying the underlying large models. This approach decouples task-specific optimization from large foundation models and enables more flexible, cost-effective deployment of LLM-powered agents in real-world applications.