cs.AIOct 11, 2025

Failure-Driven Workflow Refinement

Authors: Jusheng Zhang, Kaitong Cai, Qinglin Zeng, Ningyuan Liu, Stephen Fan, Ziliang Chen, Keze Wang

Organizations: Sun Yat-sen University · X-Era AI Lab

Abstract

Optimizing LLM-based workflows is typically formulated as a global search, where candidate workflows are evaluated based on a scalar metric. This paradigm, however, suffers from a critical flaw: information collapse. By reducing rich, multi-step execution traces to simple success/failure signals, existing methods are rendered blind to the underlying structure of failures, fundamentally preventing them from modeling the workflow's failure distribution. We reconceptualize this challenge as a distributional problem. We propose a new paradigm where the optimization goal is not to maximize a scalar score, but to directly minimize a workflow's Expected Failure Mass, i.e., the integral of its failure probability density function defined over a high-dimensional Failure Signature Space (FSS). This distributional lens allows us to move from inefficient, zero-order optimization to a principled, gradient-like descent on the failure landscape itself. We introduce CE-Graph, a framework that operationalizes this paradigm through a novel, failure-driven refinement process. CE-Graph approximates the failure distribution from a pool of counterexamples, identifies its densest regions as recurring failure modes, and applies targeted, operator-constrained graph edits via a Propose-and-Verify mechanism to greedily reduce the failure mass. On math, code, and QA benchmarks, our CE-Graph achieves higher robustness at a significantly lower cost than strong baselines. This suggests that a system's reliability emerges not from avoiding failures, but from systematically learning and reshaping the geometric structure of its failure distributions.

Figures & tables

Appendix figures & tables8 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. CB-Orchestrator: Adaptive Workflow Optimization for LLM Agents via Contextual Bandits

    Oct 9, 2026Jiahang Sun, Zhiwei Shang, Zhipiao Liu +4LLM Agent Workflow OptimizationAgentic Workflows

  2. GraphFlow: A Graph-Based Workflow Management for Efficient LLM-Agent Serving

    May 21, 2026Ao Li, Shangpeng Yang, Fahao Chen +3LLM Agent OrchestrationAgentic Workflows

  3. From Static Templates to Dynamic Runtime Graphs: A Survey of Workflow Optimization for LLM Agents

    Mar 23, 2026Ling Yue, Kushal Raj Bhandari, Ching-Yun Ko +6LLM Agent Workflow OptimizationAgentic Workflows