metaTextGrad: Automatically optimizing language model optimizers
This addresses the inefficiency of human-designed optimizers in AI systems, offering a tailored solution for tasks using LLM-based optimization, though it is incremental as it builds on frameworks like DSPy and TextGrad.
The paper tackles the problem of manually designed, general-purpose LLM-based optimizers by proposing metaTextGrad, a meta-optimizer that enhances existing optimizers for specific tasks, resulting in an average absolute performance improvement of up to 6% across benchmarks.
Large language models (LLMs) are increasingly used in learning algorithms, evaluations, and optimization tasks. Recent studies have shown that using LLM-based optimizers to automatically optimize model prompts, demonstrations, predictions themselves, or other components can significantly enhance the performance of AI systems, as demonstrated by frameworks such as DSPy and TextGrad. However, optimizers built on language models themselves are usually designed by humans with manual design choices; optimizers themselves are not optimized. Moreover, these optimizers are general purpose by design, to be useful to a broad audience, and are not tailored for specific tasks. To address these challenges, we propose metaTextGrad, which focuses on designing a meta-optimizer to further enhance existing optimizers and align them to be good optimizers for a given task. Our approach consists of two key components: a meta prompt optimizer and a meta structure optimizer. The combination of these two significantly improves performance across multiple benchmarks, achieving an average absolute performance improvement of up to 6% compared to the best baseline.