Xianzheng Ma

Language, when self-evolving and treated as a physical representation, can unlock state-of-the-art physical plausibility in video world models, even surpassing proprietary systems.

Visual intelligence could be the key to unlocking AGI, challenging the dominance of language-based models in the quest for general intelligence.

Multi-3DLLM not only excels at multi-object reasoning but also boosts performance on single-object classification tasks, revealing a surprising synergy in geometric understanding.

LVLMs can get a 16.5%-29.5% boost on visual reasoning benchmarks just by iteratively checking their work against the image, no training needed.