AI summaryⓘ
The authors study Generative Engine Optimization (GEO), which changes web pages to make them more likely to be used by AI search engines, sometimes promoting weak or false information. They create a large dataset called GEOFlagBench to test how well current methods detect GEO webpages and find existing methods have some shortcomings. To improve detection, the authors develop a new training approach called Intervention-Paired Training (IPT), which boosts accuracy considerably. They also build a system to check how trustworthy citations on GEO pages are and run their full system on real Google search results, finding that almost 9% of pages are GEO-optimized. This work lays the groundwork for better spotting and understanding GEO in search results.
Generative Engine Optimization (GEO)generative search engineswebpage optimizationGEOFlagBenchIntervention-Paired Training (IPT)GEO detectionModernBERTsource verifiabilitysearch engine auditingmachine learning benchmark
Authors
Junjie Chu, Ye Leng, Mingjie Li, Yun Shen, Xinyue Shen, Yang Zhang
Abstract
Generative Engine Optimization (GEO) modifies web content to increase its likelihood of being selected and cited by generative search engines. This can give strategically optimized pages visibility disproportionate to their authority or relevance and even make weak or false information appear well supported. Unlike conventional search, generative search synthesizes information into direct answers rather than presenting competing sources, which can further amplify these risks, as assessing source provenance and authority requires additional user interaction. Despite these concerns, systematic methods for detecting GEO-optimized webpages remain underexplored. We introduce \texttt{GEOFlagBench}, a benchmark of 3,200 webpages spanning 400 queries, four domains, and eight GEO optimizer families, and use it to systematically evaluate existing GEO detection methods. Although the strongest baseline achieves an aggregate F1 of 0.880, method-level and authorship-conditioned evaluations reveal substantial weaknesses and potential reliance on authorship-related shortcuts. We therefore propose \emph{Intervention-Paired Training} (IPT), which supervises detector responses to GEO interventions and non-GEO AI polishing; on ModernBERT, IPT improves F1 from 0.862 to 0.944 and worst-group accuracy from 0.725 to 0.883. We develop a GEO-gated Agent system for auditing the Source Tier and verifiability of Citation URLs in detected GEO pages. Finally, we deploy the complete pipeline on released Google Search and Gemini-grounded retrieval results for 1,000 real-user queries. Across 10,095 available pages, we estimate an overall GEO prevalence of 8.90\%, reaching 16.36\% among pages modified in 2026. Our results establish a foundation for systematically detecting, auditing, and measuring GEO in real-world search ecosystems.