Publications
A collection of my research work.
Narratology meets text-to-image: a survey of consistency in AI generated storybook illustrations
Zhedong Lin, Zhongsheng Wang, Qian Liu, Xinyu Zhang, Jiamou Liu
Artificial Intelligence Review 2026
Provides a comprehensive survey that grounds AI‑generated storybook illustration in narratology, defining six dimensions of consistency and reviewing methods, datasets, metrics and challenges across time, space, character, plot, style and theme.
LLM-based Business Process Models Generation from Textual Descriptions
Xiaoxuan Li, Lin Ni, Xin Wang, Tang Yitong, Ruoxuan Li, Jiamou Liu, Zhongsheng Wang
Proceedings of the 14th International Joint Conference on Natural Language Processing and the 4th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics 2025
R-Debater: Retrieval-Augmented Debate Generation through Argumentative Memory
Maoyuan Li†, Zhongsheng Wang†, Haoyuan Li, Jiamou Liu
arXiv preprint arXiv:2512.24684 2025
This article employs a multi-agent simulation approach, combined with relevant theoretical methods from debate, to implement an enhanced debate task, thereby improving the reliability of debate content.
Chatlogic: Integrating logic programming with large language models for multi-step reasoning
Zhongsheng Wang, Jiamou Liu, Qiming Bao, Hongfei Rong, Jingfeng Zhang
2024 International Joint Conference on Neural Networks (IJCNN) 2024
This paper proposes a bidirectional transformation method that converts natural language inference into logical code to enhance the multi-step reasoning ability of LLM.
Epic-Level Text Generation with LLM through Auto-prompted Reinforcement Learning
Qianqian Qi, Lin Ni, Zhongsheng Wang, Libo Zhang, Jiamou Liu, Michael Witbrock
2024 International Joint Conference on Neural Networks (IJCNN) 2024
This article implements a method for generating ultra-long novel texts, ensuring a rigorous structure and continuous content.
Cora: Optimizing low-rank adaptation with common subspace of large language models
Xiaojun Xiao, Sen Shen, Qiming Bao, Hongfei Rong, Kairui Liu, Zhongsheng Wang, Jiamou Liu
arXiv preprint arXiv:2409.02119 2024
Introduces CoRA, a method for improving Low‑Rank Adaptation (LoRA) by replacing one of its matrices with a shared subspace across tasks, reducing parameters while maintaining or improving fine‑tuning performance.
Weak Supervision Techniques towards Enhanced ASR Models in Industry-level CRM Systems
Zhongsheng Wang, Sijie Wang, Jia Wang, Yung-I Liang, Yuxi Zhang, Jiamou Liu
International Conference on Neural Information Processing 2024
Presents a weak‑supervision strategy for fine‑tuning automatic speech recognition models tailored to customer relationship management systems, leading to significant improvements in identifying customer voices in industry applications.