The advent of emerging technologies and algorithms like large language models (LLMs) are revolutionizing recommender systems, offering unprecedented opportunities to enhance personalization, interpretability, fairness, and novelty. However, integrating these technologies into recommendation workflows introduces challenges such as algorithmic bias, opaque decision-making, scalability constraints, and the need for robust evaluation frameworks. This special session seeks to bridge cutting-edge research and practical applications by fostering interdisciplinary discussions on leveraging LLMs or other emerging technologies to advance recommender systems across four pillars:

  • Personalization: Tailoring recommendations to individual user preferences while adapting to dynamic contexts.
  • Interpretability: Enabling transparent reasoning for recommendations to build user trust.
  • Fairness: Mitigating biases in content generation and user modeling.
  • Novelty: Balancing relevance with serendipity to diversify user experiences.

This session will highlight innovations in algorithms, user interaction paradigms, and evaluation methodologies, emphasizing ethical and societal implications. By uniting researchers from machine learning, HCI, ethics, and industry, we aim to define future directions for trustworthy, user-centric generative recommendation systems.

The special session invites submissions addressing (but not limited to) the following areas:

  • Improving Recommender Algorithms
    • Integrating LLMs or other emerging technologies into collaborative filtering, sequential, or causal recommendation frameworks.
    • Enhancing efficiency (e.g., model distillation, prompt engineering) for real-time LLM-driven recommendations.
    • Addressing cold-start challenges using LLM-generated synthetic data or knowledge augmentation.
    • Multi-task learning for cross-domain personalization (e.g., combining text, image, and graph data).
  • Generating Personalized Content via LLMs
    • Dynamic AIGC (AI-generated content) for adaptive item creation (e.g., personalized ads, product descriptions).
    • Controllable generation techniques to align outputs with user preferences and ethical guidelines.
    • Cross-modal personalization (e.g., LLM-driven video summaries, music playlists).
    • User-in-the-loop frameworks for co-creating or refining generative content.
  • Evolving User-System Interaction Paradigms
    • Conversational recommenders leveraging LLMs for natural, multi-turn dialogues.
    • Proactive recommendation agents that anticipate user needs through contextual reasoning.
    • Explainable interfaces using LLMs to justify recommendations in user-friendly language.
    • Multi-modal interaction (voice, text, and gestures) for inclusive and accessible systems.
  • Enhancing Trustworthiness in Recommendations
    • Bias and fairness auditing in LLM-generated content (e.g., demographic parity in job recommendations).
    • Privacy-preserving LLM architectures (e.g., federated learning, differential privacy).
    • Explainability tools to trace recommendation logic (e.g., attention visualization, counterfactual explanations).
    • Compliance frameworks for copyright, safety, and regulatory standards in generative recommendations.
  • Refining Evaluation Methodologies
    • Novel metrics for assessing creativity, serendipity, and long-term user satisfaction.
    • Human-AI collaborative evaluation (e.g., hybrid metrics combining LLM feedback and user surveys).
    • Benchmarking LLM-based recommenders against traditional models in real-world scenarios.
    • Longitudinal studies to measure societal impact (e.g., filter bubbles, diversity erosion).

Important Dates

  • Special Session Papers Submission: 15 July 2025
  • Acceptance Notification: 15 August 2025
  • Camera-Ready Submission: 08 September 2025
  • Conference Date: 16-18 Oct 2025

Paper submission instruction

Paper submission system is available at: https://cmt3.research.microsoft.com/BESC2025.

All papers will be reviewed by the Program Committee on the basis of technical quality, relevance to BESC 2025, originality, significance and clarity.

Please note:

  • Paper review will be double-blind, and submissions not properly anonymized will be desk-rejected without review.
  • Submitted papers must not substantially overlap with papers that have been published or that are simultaneously submitted to a journal or a conference with proceedings.
  • Papers must be clearly submitted in English and will be selected based on their originality, timeliness, significance, relevance, and clarity of presentation.
  • Submission of a paper should be regarded as a commitment that, should the paper be accepted, at least one of the authors will register and attend the conference to present the work.
  • Top quality papers after presented in the conference will be selected for extension and publication in several special issues of international journals, e.g., IEEE Transactions on Computational Social Systems (IEEE), IEEE Transactions on Learning Technologies (IEEE), CCF Transactions on Pervasive Computing and Interaction (Springer), EURASIP Journal on Image and Video Processing (Springer), World Wide Web Journal (Springer), Social Network Analysis and Mining (Springer), Human-Centric Intelligent Systems (Springer), Natural Language Processing (Elsevier), Health Information Science and Systems (Springer), Web Intelligence (IOS Press), etc.
  • The use of artificial intelligence (AI)–generated text in an article shall be disclosed in the acknowledgements section. The sections of the paper that use AI-generated text shall have a citation to the AI system used to generate the text.

Organizing Committee

Program Chairs
  • Assistant Prof. Nengjun Zhu, Shanghai University, China
  • Associate Prof. Qi Zhang, Tongji University, China
  • Prof. Liang Hu, Tongji University, China
  • Assistant Prof. Tianzi Zang, Nanjing University of Posts and Telecommunications, China
Program Committee Members (Tentative)
  • Chaochao Chen, Zhejiang University, China
  • Defu Lian, University of Science and Technology of China, China
  • Yuanbo Xu, Jilin University, China
  • Hui Cai, Nanjing University of Posts and Telecommunications, China
  • Tianzi Zang, Nanjing University of Posts and Telecommunications, China
  • Xinzhi Wang, Shanghai University, China
  • Hang Yu, Shanghai University, China
  • Shanshan Feng, Shandong Normal University, China
  • Jianqi Gao, Shanghai Jiao Tong University, China
  • Yong Liu, Nanyang Technological University, Singapore
  • Guanfeng Liu, Macquarie University, Australia
  • Wenpeng Lu, Qilu University of Technology, China