Samuel Dodge — accepted AI-conference papers

2024

"MM1: Methods, Analysis & Insights from Multimodal LLM Pre-training"

Brandon McKinzie, Zhe Gan, Jean-Philippe Fauconnier, Samuel Dodge, Bowen Zhang, Philipp Dufter +23

ECCV 2024poster

"In this work, we discuss building performant Multimodal Large Language Models (MLLMs). In particular, we study the importance of various architecture components and data choices. Through careful and comprehensive ablations of the image encoder, the vision language connector, and various pre-trainin…

Cited by 276 Source ↗PDF ↗code ↗Scholar ↗

2019

Finding Task-Relevant Features for Few-Shot Learning by Category Traversal

Hongyang Li, David Eigen, Samuel Dodge, Matthew Zeiler, Xiaogang Wang

CVPR 2019oral

Few-shot learning is an important area of research. Conceptually, humans are readily able to understand new concepts given just a few examples, while in more pragmatic terms, limited-example training situations are common practice. Recent effective approaches to few-shot learning employ a metric-le…

Cited by 477 PDF ↗code ↗Scholar ↗