Search papers, labs, and topics across Lattice.
University of Science & Technology of China
3
0
3
4
Knowledge transfer in MLLM fusion is not a blanket inheritance but a selective process favoring high-level reasoning over perception.
Pair-Space Generation achieves up to 4x faster generative reranking while enhancing user engagement on platforms with over 400 million daily users.
Achieving duplicate-free item selection in recommendation systems without sacrificing efficiency, DIRECTOR revolutionizes how we approach reranking by leveraging transport-optimized retrieval.
