Mengping Yang (杨孟平)

kobeshegu[at]gmail.com     Google Scholar   Github   Zhihu   CV

I recently obtained my Ph.D degree (from Sep. 2019 - Jun. 2024) from Ecust China University of Science and Technology (ECUST). My research interests mainly include multi-model learning/ AIGC, e.g., content generation of 2D images and videos, with Generative Adversarial Networks, Diffusion and auto-regressive models.

Before that, I received my B.S. from the Department of Computer Science and Technology with several honors in ECUST in Jul. 2019.

I admire distinguished researchers/engineers who promote the advancement of the community, their fascinating projects inspire me a lot! Hope that I can also make some impactful and insightful work!

I am actively looking for a long-term intern/collabration/full-time job opportunities (available from Sep. 2023), working on fundamental research and application related problems of generative models. Here is my CV, feel free to email me for any potential opportunities!

We propose an effective approach for incorporating LLMs into text-to-image diffusion models, improving the awareness of LLMs towards the CLIP visual and textual space, thus facilitating more expressive language understanding. Moreover, we devise an efficient three-stage training pipeline that accomplish fast adaptation of LLM textual features with a small amount of resources, serving as an strong baseline of integrating LLMs into diffusion models and paving the way of this important topic.

We propose an attention calibration mechanism to improve the concept-level understanding of the T2I model. Specifically, we first introduce new learnable modifiers bound with classes to capture attributes of multiple concepts. Then, the classes are separated and strengthened following the activation of the cross-attention operation, ensuring comprehensive and self-contained concepts. Additionally, we suppress the attention activation of different classes to mitigate mutual influence among concepts.

We make in-depth analyses on how to represent a data point in the feature space, how to calculate a fair distance using selected samples, and how many instances to use from each set. Together with these analysis, we build a comprehensive system for synthesis comparison, providing reliable and consistent ranks for unsupervised image generation models including GANs and Diffusion Models.

We provide a comprehensive survey on image synthesis under limited data, including data-efficient generative modeling, few-shot generative adaptation, few-shot and one-shot image synthesis.

We propose textural modulation (TexMod) and strctural discriminator (StructD) for improving the performance of few-shot image generaion.

We propose semantic-aware generator (SAG) and low-level feature augmentation (LFA) for improving the performance of few-shot image generaion.

we propose ProtoGAN, a GAN that incorporates the metric-learning-based prototype mechanism into adversarial learning by aligning the prototypes and features of synthesized distribution and the real distribution.

we propose DFSGAN for few-shot image generation, which takes dynamic Gaussian mixture (DGM) latent codes as the generator’s input.

We propose a frequency-aware model for training GANs under limited data, facilitating high-quality few-shot image syntheisi.

We propose a frequency-aware model for few-shot image generation, enabling high-fidelity synthesis for downstream tasks.

We develop Better Embedding and More Shots to address the distorted embedding of target data in few-shot learning.

The website template was adapted from Jon Barron.