Reproduce “Learning Personalized End-to-End Goal-Oriented Dialog” 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
Liangchen Luo, Wenhao Huang, Qi Zeng, Zaiqing Nie, Xu Sun
The authors introduce Personalized MemN2N, a restaurant-dialog system that uses user profiles to guide memory reading and candidate-response ranking. It combines profile embeddings, conversation histories from users with the same profile, and preferences over knowledge-base properties when the corresponding item has been mentioned. On Task 5, it reaches 95.33% per-response accuracy versus 87.28% for Split MemN2N, on this dataset.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.