AutoGEO Mini (Qwen1.7B, E-commerce)

AutoGEO Mini is a specialized GEO (Generative Engine Optimization) model developed by cx-cmu, built upon the Qwen 1.7B architecture. Its primary function is to rewrite web documents to improve their integration and visibility within answers generated by large language model-based generative engines, such as GPT, Gemini, or Claude. The model ensures that the rewritten content maintains its original meaning and factual accuracy.

Key Capabilities

• Document Rewriting: Optimizes web document text to align with the preferences of generative search engines.
• Enhanced Visibility: Aims to increase the likelihood of a document being included and covered in LLM-generated responses.
• Meaning Preservation: Designed to retain the original semantic content and factual information during the rewriting process.
• E-commerce Specialization: Specifically trained and optimized for e-commerce datasets.
• Gemini Integration: Tailored for use with the gemini-2.5-flash-lite generative engine.

Usage and Framework

This model is an integral part of the AutoGEO framework, which provides tools for document rewriting and evaluation. Developers can integrate AutoGEO Mini into their workflows using the provided Python utilities. For use with other generative engines or datasets, post-training of Qwen/Qwen3-1.7B using the AutoGEO code is recommended.

Research Context

The model is presented in the paper "What Generative Search Engines Like and How to Optimize Web Content Cooperatively" (arXiv:2510.11438), highlighting its role in optimizing web content for generative search.