7 papers
Mitigating Reward Hacking in RLHF via Bayesian Non-negative Reward Modeling
Zhibin Duan, Guowei Rong, Zhuo Li +3
Reward models learned from human preferences are central to aligning large language models (LLMs) via reinforcement learning from human feedback, yet they are often vulnerable to r…
Enhancing Uncertainty Estimation and Interpretability via Bayesian Non-negative Decision Layer
Xinyue Hu, Zhibin Duan, Bo Chen +1
Although deep neural networks have demonstrated significant success due to their powerful expressiveness, most models struggle to meet practical requirements for uncertainty estima…
Scalable Weibull Graph Attention Autoencoder for Modeling Document Networks
Chaojie Wang, Xinyang Liu, Dongsheng Wang +3
Although existing variational graph autoencoders (VGAEs) have been widely used for modeling and generating graph-structured data, most of them are still not flexible enough to appr…
A Non-negative VAE:the Generalized Gamma Belief Network
Zhibin Duan, Tiansheng Wen, Muyao Wang +2
The gamma belief network (GBN), often regarded as a deep topic model, has demonstrated its potential for uncovering multi-layer interpretable latent representations in text data. I…
Disentangled Generative Graph Representation Learning
Xinyue Hu, Zhibin Duan, Xinyang Liu +6
Recently, generative graph models have shown promising results in learning graph representations through self-supervised methods. However, most existing generative graph representa…
Beyond Spectral Decomposition: Bayesian Contrastive Learning and its Non-negative Formulation via Factor Analysis
Zhibin Duan, Tiansheng Wen, Yifei Wang +3
Factor analysis, often regarded as a Bayesian variant of matrix factorization, offers superior capabilities in capturing uncertainty, modeling complex dependencies, and ensuring ro…
Advancing Graph Generation through Beta Diffusion
Xinyang Liu, Yilin He, Bo Chen +1
Diffusion models have excelled in generating natural images and are now being adapted to a variety of data types, including graphs. However, conventional models often rely on Gauss…