DialogXpert : Driving Intelligent and Emotion-Aware Conversations through Online Value-Based Reinforcement Learning with LLM Priors

Tazeek Bin Abdur Rakib, Ambuj Mehrish, Lay-Ki Soon, Wern Han Lim, Soujanya Poria · 2025

Read DialogXpert : Driving Intelligent and Emotion-Aware Conversations through Online Value-Based Reinforcement Learning with LLM Priors by Tazeek Bin Abdur Rakib, Ambuj Mehrish, Lay-Ki Soon, Wern Han Lim, Soujanya Poria (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-17795.txt.

Description

Tazeek Bin Abdur Rakib and co-authors build a dialogue agent that plans ahead and tracks user emotions in this 2025 paper. Chatbots react well but plan poorly. Greedy word-by-word replies miss long-term goals. Their system, DialogXpert, lets a large model propose a few good moves each turn. A compact value network then picks the best move. Emotion tracking shapes each choice alongside the task goal.

Learning happens online, from the value of each action. The candidate trick keeps planning cheap and fast. Benchmarks covered negotiation, emotional support, and tutoring chats. Conversations reached their goals in very few turns in tests. Success rates are reported in the abstract's claims. The paper is a dialogue systems study.

Catalog record

CREATORTazeek Bin Abdur Rakib, Ambuj Mehrish, Lay-Ki Soon, Wern Han Lim, Soujanya Poria
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE69 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2505-17795

Browse the full library · About the Tower of Babel