# Improving Multi-Domain Task-Oriented Dialogue System with Offline Reinforcement Learning

arXiv: https://arxiv.org/abs/2411.05340 · HTML: https://mlsanity.com/paper/2411.05340/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Language-conditioned policies and VLAs, one of 13 directions in Offline RL and imitation learning from data.
- Builds on 10 of its 18 references that are in the map.
- First posted on arXiv 8 Nov 2024.

## Placed in

- Offline RL and imitation learning from data / [Language-conditioned policies and VLAs](https://mlsanity.com/offlinerl/language-conditioned-policies-and-vlas/)

## Builds on (most cited first, 10 of 10)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [A Simple Language Model for Task-Oriented Dialogue](https://mlsanity.com/paper/2005.00796/) (2020)
- [Multi-Task Pre-Training for Plug-and-Play Task-Oriented Dialogue System](https://mlsanity.com/paper/2109.14739/) (2021)
- [UBAR: Towards Fully End-to-End Task-Oriented Dialog Systems with GPT-2](https://mlsanity.com/paper/2012.03539/) (2020)
- [Task-Oriented Dialog Systems that Consider Multiple Appropriate Responses under the Same Context](https://mlsanity.com/paper/1911.10484/) (2019)
- [Conversational AI: The Science Behind the Alexa Prize](https://mlsanity.com/paper/1801.03604/) (2018)
- [A Probabilistic End-To-End Task-Oriented Dialog Model with Latent Belief States towards Semi-Supervised Learning](https://mlsanity.com/paper/2009.08115/) (2020)
- [Domain State Tracking for a Simplified Dialogue System](https://mlsanity.com/paper/2103.06648/) (2021)
- [Efficient Task-Oriented Dialogue Systems with Response Selection as an Auxiliary Task](https://mlsanity.com/paper/2208.07097/) (2022)

## Abstract (arXiv)

Task-oriented dialogue (TOD) system is designed to accomplish user-defined tasks through dialogues. The TOD system has progressed towards end-to-end modeling by leveraging pre-trained large language models. Fine-tuning the pre-trained language models using only supervised learning leads to the exposure bias and token loss problem and it deviates the models from completing the user's task. To address these issues, we propose a TOD system that leverages a unified pre-trained language model, GPT2, as a base model. It is optimized using supervised learning and reinforcement learning (RL). The issues in the TOD system are mitigated using a non-differentiable reward function. The reward is calculated using the weighted sum of the success rate and BLEU evaluation metrics. The success rate and BLEU metrics in reward calculation guide the language model for user task completion while ensuring a coherent and fluent response. Our model is acquired by fine-tuning a pre-trained model on the dialogue-session level which comprises user utterance, belief state, system act, and system response. Experimental results on MultiWOZ2.1 demonstrate that our model increases the inform rate by 1.60% and the success rate by 3.17% compared to the baseline.