📝 Publications

PersuHSG: Adaptive Persuasion Strategy Planning for Dialogue Agents Based on Hierarchical Strategy Graph //

Mengqi Chen, Bin Guo, Hao Wang, Jingqi Liu, Yan Liu, Yunji Liang, Peilin Li, Yan Pan, Zhiwen Yu

• We present PersuHSG, which conceptualizes persuasion as a tripartite framework comprising cognition, affection, and volition, with each stage represented as a graph layer and principle-based strategies for efficient multi-stage persuasion.

Neuro-Sym Supporter: A Thoughtful Emotion Support Agent Integrating Neural and Symbolic Policy Learning \

Minghui Ma, Bin Guo, Mengqi Chen, Jingqi Liu, Yasan Ding, Yan Liu, Han Wang

• We propose Sym-Mind, a differentiable logic-based reasoning framework for emotional support strategy selection, which unifies interpretability with stable performance to improve an agent’s ability to proactively explore solutions through reasoning.

Multi-agent collaborative crowd intelligence persuasive dialogue inspired by cognitive psychology \

Mengqi Chen, Bin Guo, Hao Wang, Jingqi Liu, Kaixing Zhao, Yunji Liang, Zhiwen Yu

• Making AI dialogue systems more persuasive can improve topic control and logic in conversations, enhancing human-computer interaction. We propose “Collective Intelligence Persuasion Dialogue,” in which multiple specialized AI agents collaborate.

The Future of Cognitive Strategy-enhanced Persuasive Dialogue Agents: New Perspectives and Trends

Mengqi Chen, Bin Guo, Hao Wang, Haoyu Li, Qian Zhao, Jingqi Liu, Zhiwen Yu

• Persuasion has garnered extensive attention from researchers within the field of intelligent dialogue systems. The cognitive strategy-enhanced persuasive dialogue agent (defined as CogAgent), which incorporates cognitive strategies to achieve persuasive targets through conversation, has become a predominant research paradigm.

Cascade Context-oriented Spatio-temporal Attention Network for Efficient and Fine-grained Video-grounded Dialogue
Hao Wang, Bin Guo, Mengqi Chen, Yasan Ding, Ying Zhang, Zhiwen Yu

• Video-Grounded Dialogue System (VGDS), focusing on generating reasonable responses based on multi-turn dialogue contexts and a given video, has received intensive attention recently. We propose a novel system, namely Cascade cOntext-oriented Spatial-Temporal Attention Network (COSTA), to generate reasonable responses efficiently and accurately.

Memory-enhanced Emotional Support Conversations with Motivation-driven Strategy Inference
Hao Wang, Bin Guo, Mengqi Chen, Yasan Ding, Qiuyun Zhang, Zhiwen Yu

• The prevalence of mental disorders is increasing globally, highlighting the need for effective emotional support systems. Emotional Support Conversation (ESC) systems have emerged as a promising solution, providing supportive responses to individuals experiencing emotional distress. we propose MAGIC, a Memory-enhanced emotional support conversation system with motivation-driven strAteGy InferenCe.

[Improved faster R-CNN for fabric defect detection based on Gabor filter with Genetic Algorithm optimization] ESI Hot Paper 🔥(https://dl.acm.org/doi/10.1145/3583758)
Mengqi Chen, Lingjie Yu, Chao Zhi, Runjun Sun, Shuangwu Zhu, Zhongyuan Gao, Zhenxia Ke, Mengqiu Zhu, Yuming Zhang

• Fabric defect detection plays a crucial role in fabric inspection and quality control. We improved the Faster Region-based Convolutional Neural Network (Faster R-CNN) model by embedding Gabor kernels into Faster R-CNN, termed the Genetic Algorithm Gabor Faster R-CNN (Faster GG R-CNN).