Reproduce “DialogXpert: Driving Intelligent and Emotion-Aware Conversations through Online Value-Based Reinforcement Learning with LLM Priors” 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
Tazeek Bin Abdur Rakib, Ambuj Mehrish, Lay-Ki Soon, Wern Han Lim, Soujanya Poria
DialogXpert plans goal-directed dialogue by using a frozen large language model to propose actions and a learned value model to choose among them, with inferred user emotions included in its state. In LLM-simulated evaluations across negotiation, emotional support, and tutoring, it reports 2.32 average turns and 0.9746 success on CraigslistBargain. Real-user performance is not established.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.