Hao Xu

LLMs can now resolve knowledge conflicts in real-time, leading to more reliable and interpretable outputs.

Encoder quality is capability-specific, with specialized models outperforming generic ones when pretraining aligns with task requirements.

The new REO framework reveals that the true challenge in differential equation discovery lies not just in recovering equations, but in leveraging them to reshape scientific understanding.

Forget scaling laws – this zero-shot navigation agent beats million-sample trained models by structurally unifying language, vision, and robot actions within the reasoning capabilities of pre-trained MLLMs.

MLLMs struggle to grasp the nuances of ancient Chinese script evolution, but a glyph-driven fine-tuning approach unlocks surprisingly strong performance even in smaller models.

Turns out, MLLMs struggle with manufacturing tasks not because they can't "see," but because they lack the domain-specific knowledge to understand what they're looking at.

By explicitly modeling the latent human evaluation process, VRM offers a more robust reward model, sidestepping the pitfalls of spurious correlations that plague traditional methods.

LLMs implicitly know if their reasoning steps are correct *during* generation, according to a new step-level interpretability method.