Context
Liu and Xu introduce FeatGEO, which optimizes interpretable webpage structural, content and linguistic features before generating revised text. Their GEO-Bench experiments across three generative engines report stronger citation visibility than token-level rewrite baselines while maintaining or improving content quality.
What changed
The original April 21, 2026 research preprint, later peer-reviewed and published at ACL 2026 in July, provides controlled evidence on multi-objective, feature-level GEO optimization rather than sentence-level rewriting alone.
Why it matters
AI citation visibility and content quality are not interchangeable. This work suggests document-level properties can be evaluated jointly with citation outcomes, while requiring controlled testing rather than assuming individual rewrite tactics work universally.
Visibility OS interpretation
Visibility OS interpretation: This is evidence for feature-level GEO evaluation in controlled experimental engines, not proof of how commercial production ranking or citation systems reward any given page feature.
What remains uncertain
Will these effects replicate on live search interfaces, newer model generations and natural web retrieval? How stable are the reported citation-quality trade-offs and quality judgments across domains?