Search papers, labs, and topics across Lattice.
4
0
11
7
Targeted interventions can significantly enhance the robustness of CLIP models against Typographic Attacks without additional training, outperforming existing defense methods.
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.
