The ACM Transactions on Recommender Systems (TORS) published a call for papers for its special issue Knowledge Transferring for Recommender Systems (KT4Rec).

Guest Editors:

• Zhiwei Liu, Salesforce Inc., USA
• Hao Peng, Beihang University, China
• Caiming Xiong, Salesforce Inc., USA
• Julian McAuley, University of California San Diego, USA
• Philip S. Yu, University of Illinois at Chicago, USA

The prosperity of online contents populates the widespread usage of recommender systems, including news recommendation, Ads recommendation, E-commerce recommendation, etc. However, the data scarcity problems severely degrade the recommendation ability of most existing models and impede the further development of recommender systems. Hence, we value that the knowledge transferred from other resources, for example the human knowledge from pre-trained language models, will remedy the data scarcity problem and thus improve the development of recommender systems.

However, knowledge transferring is rather challenging. In general, there is data discrepancy when transferring knowledge from source domain to target domain. Moreover, the token spaces are not unified for different recommender systems and are usually of large scales. In addition, transferring knowledge in complex scenarios of recommender systems even leads to negative impacts if without proper design. 

Therefore, raising discussions regarding the knowledge transferring in recommender systems is urgent yet just at time. Knowledge transferring in recommender systems can be from various perspectives, one domain to another, one model to another, one task to another, etc. We believe the investigation on knowledge transferring for recommender systems (KT4Rec) will advance the current recommender systems and lead to the next-generation recommendation paradigm.

Topics

Topics of interest include, but are not limited to, the following:

• Cross-domain/source Recommendation
• Multi-modal Recommendation
• Recommendation with Pre-trained Model
• Recommendation with Knowledge Graphs
• Multi-task Training for Recommender Systems 
• Self-supervised Learning for Recommender systems
• Meta-learning for Recommendation
• Knowledge Distillation for Recommendation
• Federated Recommender Systems
• Trustworthiness in Knowledge Transfer for Recommender Systems 

Important Dates

• Submissions deadline: Aug 16^th, 2023
• First-round review decisions: Nov 16^th, 2023
• Deadline for revision submissions: Feb 29^th, 2024
• Notification of final decisions: April 30^th, 2024

Submission Information

We accept research, survey, and perspective/opinion papers. Each paper should address one or more of the above-mentioned topics or be in other scopes of KT4Rec. Submissions must be prepared according to the TORS submission guidelines (https://dl.acm.org/journal/tors/author-guidelines) and must be submitted via Manuscript Central (https://mc.manuscriptcentral.com/tors). The special issue will also consider extended versions (at least 30% new content) of papers published at conferences.

For questions and further information, please contact Zhiwei Liu at zhiweiliu @ salesforce.com.