Benchmark Designers Should "Train on the Test Set" to Expose Exploitable Non-Visual Shortcuts Paper • 2511.04655 • Published 21 days ago • 7
Learning to See Before Seeing: Demystifying LLM Visual Priors from Language Pre-training Paper • 2509.26625 • Published Sep 30 • 43
Thinking in Space: How Multimodal Large Language Models See, Remember, and Recall Spaces Paper • 2412.14171 • Published Dec 18, 2024 • 24
Cambrian-1: A Fully Open, Vision-Centric Exploration of Multimodal LLMs Paper • 2406.16860 • Published Jun 24, 2024 • 63