Rui Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Rui Liu has a total of 6 co-authors. Here are the top co-authors by number of shared publications:
Fuqiang Chen
Department of Artificial Intelligence, Guangzhou Huashang College, Guangzhou, China
1 shared publication- Haotian Zhu
New York University, USA
1 shared publication - Jinming Li
Georgia Institute of Technology, USA
1 shared publication - Kangning Gao
The George Washington University, USA
1 shared publication - Xu Yan
Trine University, USA
1 shared publication - Yi Hu
University of Southern California, USA
1 shared publication
Latest Publications
- Open Access
- 22 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
A crossover strategy integrated dung beetle optimization for global optimization problems and feature selection problems
Dung Beetle Optimization (DBO) is a recent swarm intelligence optimizer inspired by the rolling, dancing, foraging, breeding, and stealing behaviors of dung beetles. Although DBO has shown promising search capability, it may suffer from diversity loss, ineffective inferior-solution utilization, and boundary-induced stagnation when solving complex optimization problems. To address these issues, this paper proposes a Crossover Strategy Integrated Dung Beetle Optimization algorithm, named CSIDBO. The proposed method incorporates three mechanisms: a horizontal crossover strategy for dimension-level information exchange, a precise elimination mechanism for replacing low-quality individuals, and a global-best-guided boundary handling strategy for correcting infeasible solutions. Moreover, a transfer-function-based binary version of CSIDBO is developed for wrapper feature selection. The performance of CSIDBO is evaluated on CEC2017, CEC2020, and CEC2022 benchmark functions and sixteen feature selection datasets. Comparative experiments, ablation studies, convergence curves, exploration–exploitation analysis, runtime comparison, and statistical tests are conducted to provide a comprehensive evaluation. Experimental results show that CSIDBO achieves competitive optimization accuracy and feature selection performance compared with classical algorithms and recent DBO variants. Nevertheless, the results also indicate that CSIDBO is not universally optimal on all functions, and its additional operators introduce extra function evaluations and runtime overhead. These limitations are discussed to provide a balanced assessment of the proposed method.
- 22 July 2026
- Published by Springer Nature in Scientific Reports
- Vol. 16 (1)
Open AccessA crossover strategy integrated dung beetle optimization for global optimization problems and feature selection problems
Dung Beetle Optimization (DBO) is a recent swarm intelligence optimizer inspired by the rolling, dancing, foraging, breeding, and stealing behaviors of dung beetles. Although DBO has shown promising search capability, it may suffer from diversity loss, ineffective inferior-solution utilization, and boundary-induced stagnation when solving complex optimization problems. To address these issues, this paper proposes a Crossover Strategy Integrated Dung Beetle Optimization algorithm, named CSIDBO. The proposed method incorporates three mechanisms: a horizontal crossover strategy for dimension-level information exchange, a precise elimination mechanism for replacing low-quality individuals, and a global-best-guided boundary handling strategy for correcting infeasible solutions. Moreover, a transfer-function-based binary version of CSIDBO is developed for wrapper feature selection. The performance of CSIDBO is evaluated on CEC2017, CEC2020, and CEC2022 benchmark functions and sixteen feature selection datasets. Comparative experiments, ablation studies, convergence curves, exploration–exploitation analysis, runtime comparison, and statistical tests are conducted to provide a comprehensive evaluation. Experimental results show that CSIDBO achieves competitive optimization accuracy and feature selection performance compared with classical algorithms and recent DBO variants. Nevertheless, the results also indicate that CSIDBO is not universally optimal on all functions, and its additional operators introduce extra function evaluations and runtime overhead. These limitations are discussed to provide a balanced assessment of the proposed method.
- Open Access
- 31 December 2025
- Published by MDPI AG in Preprints.org
Contextual Trust Evaluation for Robust Coordination in Large Language Model Multi-Agent Systems
Large Language Model (LLM)-based multi-agent systems have emerged as a promising paradigm for tackling complex tasks that exceed individual agent capabilities. However, existing approaches often suffer from coordination inefficiencies, a lack of trust mechanisms, and suboptimal role assignment strategies. This paper presents a novel trust-aware coordination framework that enhances multi-agent collaboration through dynamic role assignment and context sharing. Our framework introduces a multi-dimensional trust evaluation mechanism that continuously assesses agent reliability based on performance history, interaction quality, and behavioral consistency. The coordinator leverages these trust scores to dynamically assign roles and orchestrate agent interactions while maintaining a shared context repository for transparent information exchange. We evaluate our framework across eight diverse task scenarios with varying complexity levels, demonstrating significant improvements over baseline approaches. Experimental results show that our trust-aware framework achieves a 87.4% task success rate, reducing execution time by 36.3% compared to non-trust-based methods, while maintaining 43.2% lower communication overhead. The framework's ability to adapt agent roles based on evolving trust scores enables more efficient resource utilization and robust fault tolerance in dynamic multi-agent environments.
- 31 December 2025
- Published by MDPI AG in Preprints.org
Open AccessContextual Trust Evaluation for Robust Coordination in Large Language Model Multi-Agent Systems
Large Language Model (LLM)-based multi-agent systems have emerged as a promising paradigm for tackling complex tasks that exceed individual agent capabilities. However, existing approaches often suffer from coordination inefficiencies, a lack of trust mechanisms, and suboptimal role assignment strategies. This paper presents a novel trust-aware coordination framework that enhances multi-agent collaboration through dynamic role assignment and context sharing. Our framework introduces a multi-dimensional trust evaluation mechanism that continuously assesses agent reliability based on performance history, interaction quality, and behavioral consistency. The coordinator leverages these trust scores to dynamically assign roles and orchestrate agent interactions while maintaining a shared context repository for transparent information exchange. We evaluate our framework across eight diverse task scenarios with varying complexity levels, demonstrating significant improvements over baseline approaches. Experimental results show that our trust-aware framework achieves a 87.4% task success rate, reducing execution time by 36.3% compared to non-trust-based methods, while maintaining 43.2% lower communication overhead. The framework's ability to adapt agent roles based on evolving trust scores enables more efficient resource utilization and robust fault tolerance in dynamic multi-agent environments.