IF-GEO: Conflict-Aware Instruction Fusion for Multi-Query Generative Engine Optimization
This work addresses the problem of source visibility in generative engines for information retrieval, offering a practical solution for content optimization, though it appears incremental as it builds on existing GEO strategies.
The paper tackles the challenge of optimizing documents for diverse queries in Generative Engine Optimization (GEO) by proposing IF-GEO, a framework that mines optimization preferences and synthesizes a revision blueprint to handle conflicting requirements, achieving substantial performance gains in experiments.
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.