Xuan Lu

The paper proposes a vision-first formulation to improve cache reuse efficiency and reranking performance in multimodal large language models.

This paper proposes CompRank, a token-efficient reranking framework for large language models that reduces redundant computation and achieves strong reranking performance.

This paper proposes an adaptive streaming reasoning framework that enables models to reason during input streaming and perform final deliberation once the stream is complete.

This paper proposes ShopX, a model-centric framework for intent-driven shopping experiences using a single foundation model for intent understanding, execution planning, and item-space operations.