Why Search When You Can Transfer? Amortized Agentic Workflow Design from Structural Priors

TL;DR

SWIFT is a framework that leverages structural priors and few-shot transfer to synthesize workflows for unseen tasks, significantly reducing computational costs compared to traditional search-based methods.

Contribution

The paper introduces SWIFT, a novel approach that amortizes workflow design into reusable priors, enabling efficient, zero-search synthesis of workflows across diverse tasks and models.

Findings

SWIFT outperforms state-of-the-art search-based methods on five benchmarks.

It reduces per-task optimization cost by three orders of magnitude.

Workflow topologies transfer effectively even with surface semantics replaced.

Abstract

Automated agentic workflow design currently relies on per-task iterative search, which is computationally prohibitive and fails to reuse structural knowledge across tasks. We observe that optimized workflows converge to a small family of domain-specific topologies, suggesting that this combinatorial search is largely redundant. Building on this insight, we propose SWIFT (Synthesizing Workflows via Few-shot Transfer), a framework that amortizes workflow design into reusable structural priors. SWIFT first distills compositional heuristics and output-interface contracts from contrastive analysis of prior search trajectories across source tasks. At inference time, it conditions a single LLM generation pass on these priors together with cross-task workflow demonstrations to synthesize a complete, executable workflow for an unseen target task, bypassing iterative search entirely. On five benchmarks, SWIFT outperforms the state-of-the-art search-based method while reducing marginal per-task optimization cost by three orders of magnitude. It further generalizes to four additional unseen benchmarks and transfers successfully from GPT-4o-mini to three additional foundation models (Grok, Qwen, Gemma). Controlled ablations reveal that workflow demonstrations primarily transfer topological structure rather than surface semantics: replacing all operator names with random strings still retains over 93% of the full system's average performance.