Prof. Madina Mansurova
Al-Farabi Kazakh National University
Kazakhstan
madina.mansurova@kaznu.kz

Prof. Peng Wang
Northwestern Polytechnical University
China
peng.wang@nwpu.edu.cn

The rapid development of Large Language Models (LLMs) and Vision-Language Models (VLMs) is transforming the landscape of Artificial Intelligence, enabling intelligent systems to process and reason across text and visual data modalities. The LVLMS 2026 Special Session at the 18th International Conference on Computational Collective Intelligence (ICCCI 2026) is devoted to discussing the recent advances, architectural innovations, and ethical considerations of integrating these advanced models into complex intelligent systems.
We want to offer an opportunity for researchers and practitioners to identify new promising research directions as well as to publish recent advances in this area.

The scope of the session includes, but is not limited to, the following topics:

• Large Language Models (LLMs) architectures and optimization techniques
• Vision-Language Models (VLMs) for multimodal reasoning
• LLMs and VLMs for Natural Language Understanding (NLU) and Generation (NLG)
• Integration of LLMs/VLMs in autonomous and intelligent systems
• Multimodal learning and cross-modal representation learning
• Fine-tuning, transfer learning, and domain adaptation of LLMs and VLMs
• Knowledge augmentation and retrieval-augmented LLMs
• Explainability and interpretability of LLMs and VLMs
• Ethical, privacy, and fairness considerations in large-scale models
• LLMs and VLMs for human-AI interaction and collaborative intelligence
• Real-time deployment of LLMs/VLMs in edge and cloud computing
• Robustness, security, and adversarial attacks on LLMs/VLMs
• Applications of LLMs/VLMs in education, healthcare, and industry
• Multilingual and cross-cultural LLMs for global AI systems
• Benchmarking, evaluation metrics, and performance analysis of LLMs and VLMs
• Hybrid AI systems combining symbolic reasoning and LLM/VLM approaches
• Dataset collection, preprocessing, and synthetic data generation for LLMs/VLMs
• Large-scale training strategies and efficiency optimization
• Collaborative and distributed learning for LLMs and VLMs
• Emerging trends and future directions in LLMs and VLMs research