IF-GEO: Conflict-Aware Instruction Fusion for Multi-Query Generative Engine Optimization

TL;DR

IF-GEO introduces a conflict-aware instruction fusion framework that optimizes multi-query content revisions for generative engines, improving source visibility and robustness across diverse retrieval scenarios.

Contribution

The paper proposes a novel diverge-then-converge framework with conflict-aware instruction fusion for multi-query optimization in generative engines, addressing conflicting revision requirements.

Findings

Significant performance improvements on multi-query benchmarks

Enhanced robustness across diverse retrieval scenarios

Effective conflict resolution in content revision

Abstract

As Generative Engines revolutionize information retrieval by synthesizing direct answers from retrieved sources, ensuring source visibility becomes a significant challenge. Improving it through targeted content revisions is a practical strategy termed Generative Engine Optimization (GEO). However, optimizing a document for diverse queries presents a constrained optimization challenge where heterogeneous queries often impose conflicting and competing revision requirements under a limited content budget. To address this challenge, we propose IF-GEO, a "diverge-then-converge" framework comprising two phases: (i) mining distinct optimization preferences from representative latent queries; (ii) synthesizing a Global Revision Blueprint for guided editing by coordinating preferences via conflict-aware instruction fusion. To explicitly quantify IF-GEO's objective of cross-query stability, we introduce risk-aware stability metrics. Experiments on multi-query benchmarks demonstrate that IF-GEO achieves substantial performance gains while maintaining robustness across diverse retrieval scenarios.