Wei Peng

7

Papers

61

Total Citations

Papers (7)

Semantics-Adaptive Activation Intervention for LLMs via Dynamic Steering Vectors

Self-Distillation Regularized Connectionist Temporal Classification Loss for Text Recognition: A Simple Yet Effective Approach

Align Before Adapt: Leveraging Entity-to-Region Alignments for Generalizable Video Action Recognition

FreeNet: Liberating Depth-Wise Separable Operations for Building Faster Mobile Vision Architectures

MorphVAE: Advancing Morphological Design of Voxel-Based Soft Robots with Variational Autoencoders

Learning from Failure: Improving Meeting Summarization without Good Samples

Diffusion Sampling Correction via Approximately 10 Parameters