BayesFlow: A Probability Inference Framework for Meta-Agent Assisted Workflow Generation

TL;DR

BayesFlow introduces a Bayesian inference-based framework for automatic workflow generation, significantly improving accuracy over existing methods by sampling workflows with theoretical guarantees and a training-free approach.

Contribution

The paper presents BayesFlow, a novel, training-free Bayesian workflow generation algorithm that outperforms state-of-the-art baselines across multiple benchmarks.

Findings

BayesFlow improves accuracy by up to 9 percentage points over SOTA.

It achieves up to 65 percentage points better than zero-shot prompting.

The framework is theoretically grounded with proven convergence properties.

Abstract

Automatic workflow generation is the process of automatically synthesizing sequences of LLM calls, tool invocations, and post-processing steps for complex end-to-end tasks. Most prior methods cast this task as an optimization problem with limited theoretical grounding. We propose to cast workflow generation as Bayesian inference over a posterior distribution on workflows, and introduce \textbf{Bayesian Workflow Generation (BWG)}, a sampling framework that builds workflows step-by-step using parallel look-ahead rollouts for importance weighting and a sequential in-loop refiner for pool-wide improvements. We prove that, without the refiner, the weighted empirical distribution converges to the target posterior. We instantiate BWG as \textbf{BayesFlow}, a training-free algorithm for workflow construction. Across six benchmark datasets, BayesFlow improves accuracy by up to 9 percentage points over SOTA workflow generation baselines and by up to 65 percentage points over zero-shot prompting, establishing BWG as a principled upgrade to search-based workflow design. Code will be available on https://github.com/BoYuanVisionary/BayesFlow.