Andi Han

Turning Internal Gap into Self-Improvement: Promoting the Generation-Understanding Unification in MLLMs

Memory-Efficient LLM Pretraining via Minimalist Optimizer Design

Consistency Is Not Always Correct: Towards Understanding the Role of Exploration in Post-Training Reasoning

ATOM: A Pretrained Neural Operator for Multitask Molecular Dynamics

Accelerating Discrete Diffusion Models with Parallel Sampling

Expanding the Chaos: Neural Operator for Stochastic (Partial) Differential Equations

Provable Benefit of Curriculum in Transformer Tree-Reasoning Post-Training

On the Optimization and Generalization of Two-layer Transformers with Sign Gradient Descent

ACT as Human: Multimodal Large Language Model Data Annotation with Critical Thinking

Generalization Bound of Gradient Flow through Training Trajectory and Data-dependent Kernel

How Does Label Noise Gradient Descent Improve Generalization in the Low SNR Regime?

Efficient Optimization with Orthogonality Constraint: a Randomized Riemannian Submanifold Method

Can Diffusion Models Learn Hidden Inter-Feature Rules Behind Images?

When Graph Neural Networks Meet Dynamic Mode Decomposition

On the Role of Label Noise in the Feature Learning Process

Provable In-Context Vector Arithmetic via Retrieving Task Concepts

On the Feature Learning in Diffusion Models

Efficient optimization with orthogonality constraint: a randomized Riemannian submanifold method

Diffusing to the Top: Boost Graph Neural Networks with Minimal Hyperparameter Tuning

Label Noise Gradient Descent Improves Generalization in the Low SNR Regime

The Role of Label Noise in the Feature Learning Process