Search papers, labs, and topics across Lattice.
Peipei Zhou and Zhuoping Yang are with Brown University. Zheng Dong is with Wayne State University. Insup Lee is with University of Pennsylvania Aidong Zhang is with the University of Virginia Robert Dick is with the University of Michigan. Majid Sarrafzadeh is with UCLA. Xiaodong Wu is with the University of Iowa Weisong Shi is with the University of Delaware Jingtong Hu is with the University of Pittsburgh Yiyu Shi is with the University of Notre Dame 1 Corresponding author: Peipei Zhou (peipei_zhou@brown.edu)
3
0
10
7
Counterfactual examples supercharge visual in-context learning, enabling smaller vision-language models to outperform larger ones by focusing on causal relationships rather than superficial correlations.
A 3B parameter model, Med-V1, matches the evidence attribution performance of GPT-5 in biomedical contexts, offering a scalable alternative to frontier LLMs.
Achieve GAM-level interpretability with deep learning accuracy by dynamically gating feature interactions through a mixture of experts.
