Shenglai Zeng, Tianqi Zheng, Chuan Tian, Dante Everaert, Yau-Shian Wang, Yupin Huang, Michael J. Morais, Rohit Patki, Jinjin Tian, Xinnan Dai, Kai Guo, Monica Xiao Cheng, Hui Liu. Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2026.

Fibrosis is a key component in the pathogenic mechanism of a variety of diseases. These diseases involving fibrosis may share common mechanisms and therapeutic targets, and therefore common intervention strategies and medicines may be applicable for these diseases. For this reason, deliberately introducing anti-fibrosis characteristics into predictive modeling may lead to more success in drug repositioning. In this study, anti-fibrosis knowledge base was first built by collecting data from multiple resources. Both structural and biological profiles were then derived from the knowledge base and used for constructing machine learning models including Structural Profile Prediction Model (SPPM) and Biological Profile Prediction Model (BPPM). Three external public data sets were employed for validation purpose and further exploration of potential repositioning drugs in wider chemical space. The resulting SPPM and BPPM models achieve area under the receiver operating characteristic curve (area under the curve) of 0.879 and 0.972 in the training set, and 0.814 and 0.874 in the testing set. Additionally, our results also demonstrate that substantial amount of multi-targeting natural products possess notable anti-fibrosis characteristics and might serve as encouraging candidates in fibrosis treatment and drug repositioning. To leverage our methodology and findings, we developed repositioning prediction platform, drug repositioning based on anti-fibrosis characteristic that is freely accessible via https://www.biosino.org/drafc.

We propose a pose-guided approach for appearance preserving video prediction by combining global and local information using Generative Adversarial Networks (GANs). The aim is to predict the subsequent frames based on previous frames of human action videos. Considering that human action videos contain both background scenes which are relatively time-invariant among frames, and human actions which are time-varying components, we use a global GAN to model the time-invariant background and coarse human profiles. Then, a local GAN is utilized to further refine the time-varying human parts. Finally, we use a 3D auto-encoder to fine-tune the frame-by-frame images to obtain the whole predicted video. We evaluate our model on the Penn Action and J-HMDB datasets and demonstrate the superiority of our proposed method over other state-of-the-art methods.

Paraphrase identification is a fundamental task in natural language process areas. During the process of fulfilling this challenge, different features are exploited. Semantically equivalence and syntactic similarity are of the most importance. Apart from advance feature extraction, deep learning based models are also proven their promising in natural language process jobs. As a result in this research, we adopted an interactive representation to modelling the relationship between two sentences not only on word level, but also on phrase and sentence level by employing convolution neural network to conduct paraphrase identification by using semantic and syntactic features at the same time. The experimental study on commonly used MSRP has shown the proposed method's promising potential.