Tianyu Wang

This work studies open-vocabulary task-oriented dexterous grasp generation, where a robot must infer functional intent from free-form language, ground it in multi-view visual observations and object geometry, and generate an executable high-degree-of-freedom grasp.

Standard SGD can get arbitrarily close to an anytime convergence rate of $\sqrt{\log n / n}$, but hitting that benchmark exactly is mathematically impossible under any deterministic step-size schedule.

AutoConcept transforms metadata into interpretable memory, significantly boosting image retrieval accuracy without the need for additional training.

EvoWiki achieves a remarkable 9.72% improvement in accuracy over existing methods, revolutionizing how we handle knowledge evolution in collaborative environments.