Min Chen

Forget sparse rewards, GSDrive unlocks dense, physics-based reward shaping for autonomous driving by rendering the environment with differentiable 3D Gaussian Splatting.

Preference optimization objectives, despite their diversity, can be steered towards disentangled dynamics that avoid suppressing the chosen response alongside the rejected one, simply by satisfying a "disentanglement band" condition.

Ditch the slow per-scene optimization: SurfelSplat reconstructs surfaces from sparse views in under a second, matching state-of-the-art accuracy with a 100x speedup.

LLMs can selectively forget specific knowledge without sacrificing overall performance, thanks to a new representation-deviation technique that precisely targets and disrupts unwanted associations.