记忆增强与多智能体协同的生成式引擎优化方法及系统
By employing a generative engine optimization method that combines memory enhancement with multi-agent collaboration, we have solved the problems of difficulty in quantifying causal effects and content distortion in generative engine optimization. This method achieves high exposure and high fidelity optimization results and possesses lifelong learning and engine adaptability.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- HANGZHOU DIANZI UNIV
- Filing Date
- 2026-03-18
- Publication Date
- 2026-07-17
AI Technical Summary
Existing generative engine optimization methods cannot scientifically quantify causal effects, are prone to content distortion and catastrophic forgetting of strategies, lack memory mechanisms, and cannot adapt to the differentiated preferences of different engines.
By employing memory enhancement and multi-agent collaboration, a two-branch controlled experimental environment is constructed. The content is optimized using multi-agent co-evolutionary mechanisms and two-layer memory mechanisms. Combined with the DSV-CF dual-axis evaluation system, lifelong learning and adaptive optimization of the strategy are achieved.
It significantly improves the overall impact and attribution accuracy of generative answers, ensuring that the optimized results have both high exposure and high credibility, and possess lifelong learning capabilities and engine preference adaptability.
Smart Images
Figure CN121859960B_ABST