| Feb 2026 | Two papers accepted to CVPR 2026. One paper introduces a new approach for understanding and retrieving images with multiple meanings. The other develops new techniques to prevent misuse of models for harmful tasks. |
| Jan 2026 | Received a NIFA funded AgriProspects grant for developing AI‑enhanced workforce training. Thanks to the AgriProspects team and the Extension Foundation for supporting this work. |
| Dec 2025 | Excited to co-organize the first Generative AI for XR and Identity-based Applications (GenXR-ID) workshop at CVPR 2026. Our workshop brings together work at the intersection of multimodality, generative models, extended reality, and biometrics. |
| Nov 2025 | Our paper on a new type of multi-image adversarial attack on multimodal large language models was accepted to AAAI 2026. |
| Aug 2025 | Two papers accepted to EMNLP 2025 main. One paper introduces a new architecture for flexible-length discrete diffusion LLM infilling, and our other paper tackles question answering bias in VLMs. We also had two Findings papers accepted, which introduce new methods for steering and fine-grained classification in VLMs. |
| Jun 2025 | We received a research grant from the Commonwealth Cyber Initiative for an exciting project related to protecting embodied agents against adversarial attacks. Thanks CCI! |
| May 2025 | Received a Google Research Scholar award for a project on making multimodal web agents safer. Thanks Google! |