5 papers
UniSteer: Text-Guided Flow Matching in Activation Space for Versatile LLM Steering
Yingdong Shi, Ruiming Zhang, Changming Li +4
Activation-based control steers large language models (LLMs) by intervening on their internal representations during inference, and has emerged as an effective paradigm for control…
Discovering Influential Neuron Path in Vision Transformers
Yifan Wang, Yifei Liu, Yingdong Shi +5
Vision Transformer models exhibit immense power yet remain opaque to human understanding, posing challenges and risks for practical applications. While prior research has attempted…
MouseGPT: A Large-scale Vision-Language Model for Mouse Behavior Analysis
Teng Xu, Taotao Zhou, Youjia Wang +12
Analyzing animal behavior is crucial in advancing neuroscience, yet quantifying and deciphering its intricate dynamics remains a significant challenge. Traditional machine vision a…
Dissecting and Mitigating Diffusion Bias via Mechanistic Interpretability
Yingdong Shi, Changming Li, Yifan Wang +5
Diffusion models have demonstrated impressive capabilities in synthesizing diverse content. However, despite their high-quality outputs, these models often perpetuate social biases…
DreamPrinting: Volumetric Printing Primitives for High-Fidelity 3D Printing
Youjia Wang, Ruixiang Cao, Teng Xu +4
Translating the rich visual fidelity of volumetric rendering techniques into physically realizable 3D prints remains an open challenge. We introduce DreamPrinting, a novel pipeline…