.png){kind=icon}
返回
ChatDiff: A ChatGPT-based diffusion model for long-tailed classification
.png){kind=icon}
PRE
AI
DOI：10.1016/j.neunet.2024.106794.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Long-tailed data distributions have been a major challenge for the practical application of deep learning. Information augmentation intends to expand the long-tailed data into uniform distribution, which provides a feasible way to mitigate the data starvation of underrepresented classes. However, most existing augmentation methods face two significant challenges: (1) limited diversity in generated samples, and (2) the adverse effect of generated negative samples on downstream classification performance. In this paper, we propose a novel information augmentation method, named ChatDiff, to provide diverse positive samples for underrepresented classes, and eliminate generated negative samples. Specifically, we start with a prompt template to extract textual prior knowledge from the ChatGPT-3.5 model, enhancing the feature space for underrepresented classes. Then using this prior knowledge, a conditional diffusion model generates semantic-rich image samples for tail classes. Moreover, the proposed ChatDiff leverages a CLIP-based discriminator to screen and remove generated negative samples. This process avoids neural network learning the invalid or erroneous features, and further, improves long-tailed classification performance. Comprehensive experiments conducted on long-tailed benchmarks such as CIFAR10-LT, CIFAR100-LT, ImageNet-LT, and iNaturalist 2018, validate the effectiveness of our ChatDiff method.
Keyword:
Long-tailed learning
Information augmentation
ChatGPT-3.5
Diffusion probabilistic model
Discriminator mechanism
Image classification
期刊
IF:
6.3
论文数:
8.2K
被引数:
3.0W
机构
引用论文
Expression of the low‐affinity NGF receptor during human muscle development, regeneration, and in tissue culture低亲和力NGF受体在人肌肉发育，再生和组织培养中的表达
PREAI
分享收藏
Sensorless Control of Z Source Inverter fed BLDC Motor Drive by FOC - DTC Hybrid Control Strategy Using Fuzzy Logic Controller采用模糊逻辑控制器的foc-dtc混合控制策略的Z源逆变器馈电BLDC电机驱动的无传感器控制
PREAI
分享收藏Bioactive compounds, antioxidant capacity and anti-inflammatory activity of native fruits from Brazil.png)
PLOS ONE
OAAI
PLOS ONEIF0
OAAI分享收藏
查看更多内容