Reproduce “A Survey on Conversational Recommender Systems” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Dietmar Jannach, Ahtsham Manzoor, Wanling Cai, Li Chen
Researchers from the University of Klagenfurt and Hong Kong Baptist University surveyed 121 papers on Conversational Recommender Systems (CRS), providing a structured overview of their interaction modalities, underlying knowledge, computational tasks, and evaluation methods. This work defines a conceptual architecture for CRS and identifies open challenges for future research.
Key insights
- CRS enable richer preference elicitation and dynamic interaction compared to traditional systems, offering explanations and accommodating user feedback through multi-turn dialogue.
- Significant advancements in natural language processing (NLP), voice-controlled assistants, and chatbot technology have made sophisticated conversational interactions feasible for recommender systems.
- The field lacks a widely adopted standard evaluation framework for CRS, with current methods facing challenges in reflecting real-world user behavior and comprehensively assessing conversation quality alongside recommendation effectiveness.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.