RA-MoWE: Workflow-Affinity Embeddings for Query Clustering and Agentic Workflow Generation
Withdrawn from ICLR 2027, so left out of the paper lists and the map.
Abstract
Agentic workflows enable large language models (LLMs) to solve complex tasks by coordinating reasoning, tool use, and verification. However, a workflow optimized for an entire task collection can overlook differences in the reasoning strategies that individual queries need, while searching for a new workflow for every query repeats costly optimization. To address this tradeoff, we introduce RA-MoWE, a framework that uses workflow-affinity embeddings to cluster queries and guide the generation of reusable expert workflows. Each embedding records how well a fixed set of reference workflows solves a query, revealing similarities in which reasoning strategies are effective. RA-MoWE uses each cluster’s queries and average embedding to initialize and refine a specialized workflow through execution feedback. An embedding encoder predicts these embeddings from query text, allowing new queries to select a generated expert without first executing the reference workflows. With predicted affinities, RA-MoWE achieves 42.90% on MixBench-H and 61.7% on MATH, using 1.98 language-model calls per MixBench-H query. With measured affinities, it reaches 62.1% on MATH and resolves 24.50% of SWE-bench Verified issues, indicating setting-dependent benefits from reusable expert workflows