Can I read DVGAN: Minimax Game for Search Diversification on EtoBox?
DVGAN: Minimax Game for Search Diversification by Xiaojie Wang is a document available to read on EtoBox.
What is DVGAN: Minimax Game for Search Diversification about?
The paper presents DVGAN, a novel framework for search result diversification that utilizes a Generative Adversarial Network (GAN) to improve the ranking of documents by combining explicit and implicit features. The framework addresses the challenge of training diversification models by generating high-quality negative samples and optimizing the diversification score function. Experimental results demonstrate that DVGAN significantly outperforms existing diversification methods on the TREC Web Track dataset
- Author
- Xiaojie Wang
- Language
- EN