NASA’s new dark energy space telescope can also detect killer asteroids
technologyreview.com·2h ago
AI Research
arXiv CS.AI·May 26·5 min read
What Gets Cited: Competitive GEO in AI Answer Engines
TL;DR
AI answer engines often cite only a few sources, impacting visibility. Researchers developed a controlled testbed for Competitive Generative Engine Optimization (GEO) to analyze citation behavior among AI-generated answers.
✦ Why It Matters
Engineers can optimize content for AI answer engines by focusing on relevance and strategic source positioning.
Key Takeaways
- Prioritize topical relevance in your content to enhance citation chances.
- Position your sources strategically in lists to improve visibility.
- Incorporate explicit price information to attract citations.
- Use recent timestamps to increase the likelihood of being cited.
How It Works
The study employs a controlled retrieval-augmented generation (RAG) testbed, where two candidate sources are injected into the model context. By measuring which source receives the first citation marker, the researchers can isolate the effects of various content factors, such as relevance and position, while controlling for biases.