FlowBot: Inducing LLM Workflows with Bilevel Optimization and Textual Gradients

TL;DR

FlowBot introduces a data-driven bilevel optimization method using textual gradients to automatically induce effective LLM workflows, reducing reliance on human-crafted pipelines.

Contribution

The paper presents a novel bilevel optimization framework with textual gradients for automatic LLM workflow induction, outperforming existing manual or generated workflows.

Findings

FlowBot achieves competitive performance against strong baselines.

The bilevel optimization approach effectively structures LLM calls.

Textual gradients enable modular optimization of LLM components.

Abstract

LLM workflows, which coordinate structured calls to individual LLMs/agents to achieve a particular goal, offer a promising path towards building powerful AI systems that can tackle diverse tasks. However, existing approaches for building such workflows generally rely on human-crafted pipelines and prompts, which presents a substantial bottleneck in real world deployment. How can we automatically induce LLM-based agents and workflows in a data-driven way? This paper describes a simple data-driven approach for automatically inducing agents and LLM workflows. We formulate workflow induction as a bilevel optimization problem: an outer loop which optimizes a high-level sketch of the workflow (in particular how the LLM calls should be structured), and an inner loop which optimizes each individual LLM call one-by one. Both loops are optimized with ``textual gradients'' where for the inner loop we optimize each component in a modular way through ``backpropagating'' textual gradients layer-by-layer. We find that LLM workflows discovered through our \textsc{FlowBot} (work\textbf{flow} induction through \textbf{b}ilevel \textbf{o}ptimization and \textbf{t}extual gradients) approach performs competitively against strong baselines that make use of human-crafted or generated workflows.