Rui Li

1/ BayesVLM turns pre-trained VLMs into probabilistic models post hoc, without retraining or fine-tuning. It is model-agnostic, works with standard #VLM architectures, and provides uncertainty estimates with virtually no inference overhead.

@natolambert Hi Nathan, thanks for sharing the opportunity. I was wondering is the applications reviewed on a rolling basis or? I applied earlier but haven't heard anything back, so I just wanted to check

We're starting to hire for our 2026 Olmo interns! Looking for excellent students to do research to help build our best models (primarily enrolled in Ph.D. with experience or interest in any area of the language modeling pipeline).

📢📢New #SpecialIssue "Advances in Probabilistic #MachineLearning" edited by Martin Trapp and Pierre Alquier, is Open for Submission @PierreAlquier 👉Submit to the Special Issue: mdpi.com/journal/entrop… 📆Submission deadline: 9 March 2025

.@ruili_pml is presenting our work 'A look at improving robustness in visual-inertial SLAM by moment matching' at #FUSION2022 in Linköping on Wednesday. The paper takes an empirical view into inference in VIO/SLAM with toy (example below) and real data. 📄 arxiv.org/abs/2205.13821