Feng Zhang

A novel semi-tensor product for third-order tensors under the t-product framework induced by arbitrary invertible linear transforms is introduced, yielding an accelerated multi-term randomized semi-tensor product SVD algorithm that achieves a balance between reconstruction accuracy and computational efficiency.

Decoupling task synthesis from on-policy rollouts solves the learning-signal saturation bottleneck in terminal agents, boosting long-horizon RL performance by up to 18 percentage points on Terminal-Bench 2.1.

NGS-Marker effectively thwarts partial infringement in 3D Gaussian Splatting, ensuring robust copyright protection for 3D assets.

Retaining more copies of low-frequency content while aggressively pruning high-frequency duplicates can significantly boost model performance during pretraining.

Trajectory neglect in LLM agents can be significantly reduced using a novel reward mechanism that enhances focus on task goals without sacrificing training stability.

LLMs can now be trained to prioritize task constraints intrinsically, resulting in a dramatic improvement in planning reliability.

A new tensor completion method achieves superior performance by leveraging a flexible nonconvex surrogate that accurately captures tensor structure.