Think Before Writing: Feature-Level Multi-Objective Optimization for Generative Citation Visibility

2026-04-21

AI Summary

This work addresses the challenge that generative answer engines, by selectively citing sources to determine content visibility, render traditional search engine optimization ineffective. Existing generative engine optimization (GEO) approaches rely on token-level rewriting, lacking interpretability and struggling to balance citation visibility with content quality. To overcome these limitations, we propose FeatGEO, a novel framework that introduces feature-level multi-objective optimization into GEO for the first time. FeatGEO abstracts webpages into interpretable attributes—such as structural, content-based, and linguistic features—and performs high-level optimization in the feature space before leveraging a language model to generate fluent text, thereby decoupling optimization from generation. Experiments on GEO-Bench demonstrate that FeatGEO significantly improves citation visibility across three major generative engines while maintaining or enhancing content quality, substantially outperforming token-level baselines and exhibiting strong generalization across models of varying scales.

Technology Category

• Search and Optimization: Learning to Search
• Natural Language Processing: Generation
• Machine Learning: Feature Construction/Reformulation

Application Category

• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Generative answer engines expose content through selective citation rather than ranked retrieval, fundamentally altering how visibility is determined. This shift calls for new optimization methods beyond traditional search engine optimization. Existing generative engine optimization (GEO) approaches primarily rely on token-level text rewriting, offering limited interpretability and weak control over the trade-off between citation visibility and content quality. We propose FeatGEO, a feature-level, multi-objective optimization framework that abstracts webpages into interpretable structural, content, and linguistic properties. Instead of directly editing text, FeatGEO optimizes over this feature space and uses a language model to realize feature configurations into natural language, decoupling high-level optimization from surface-level generation. Experiments on GEO-Bench across three generative engines demonstrate that FeatGEO consistently improves citation visibility while maintaining or improving content quality, substantially outperforming token-level baselines. Further analyses show that citation behavior is more strongly influenced by document-level content properties than by isolated lexical edits, and that the learned feature configurations generalize across language models of different scales.

Problem

• generative engine optimization
• citation visibility
• content quality
• feature-level optimization
• multi-objective optimization

Innovation

• feature-level optimization
• generative engine optimization
• multi-objective optimization
• citation visibility
• interpretable features