NEURAL REASONING LAB

NeuRLab aims to address key challenges in MLLMs. Our research spans a broad spectrum—from vision, one of the most essential modalities, to NLP, the foundation of reasoning in LLMs, and the unifying principles of deep learning. We focus on efficient MLLM architecture design and training optimization strategies, ultimately striving for reliable reasoning and progress toward AGI.