# Collaborative Multi-Agent Dialogue Model Training Via Reinforcement Learning

arXiv: https://arxiv.org/abs/1907.05507 · HTML: https://mlsanity.com/paper/1907.05507/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 11 Jul 2019.

## Built on by (most cited first, 9 of 9)

- [Controllable Dialogue Simulation with In-Context Learning](https://mlsanity.com/paper/2210.04185/) (2022)
- [Transferable Dialogue Systems and User Simulators](https://mlsanity.com/paper/2107.11904/) (2021)
- [Survey on reinforcement learning for language processing](https://mlsanity.com/paper/2104.05565/) (2021)
- [A Survey on Recent Advances and Challenges in Reinforcement Learning Methods for Task-Oriented Dialogue Policy Learning](https://mlsanity.com/paper/2202.13675/) (2022)
- [Metaphorical User Simulators for Evaluating Task-oriented Dialogue Systems](https://mlsanity.com/paper/2204.00763/) (2022)
- [A Generative User Simulator with GPT-based Architecture and Goal State Tracking for Reinforced Multi-Domain Dialog Systems](https://mlsanity.com/paper/2210.08692/) (2022)
- [Knowledge-Grounded Conversational Data Augmentation with Generative Conversational Networks](https://mlsanity.com/paper/2207.11363/) (2022)
- [Jointly Reinforced User Simulator and Task-oriented Dialog System with Simplified Generative Architecture](https://mlsanity.com/paper/2210.06706/) (2022)
- [Epistemoverse: Toward an AI-Driven Knowledge Metaverse for Intellectual Heritage Preservation](https://mlsanity.com/paper/2512.12201/) (2025)

## Abstract (arXiv)

We present the first complete attempt at concurrently training conversational agents that communicate only via self-generated language. Using DSTC2 as seed data, we trained natural language understanding (NLU) and generation (NLG) networks for each agent and let the agents interact online. We model the interaction as a stochastic collaborative game where each agent (player) has a role ("assistant", "tourist", "eater", etc.) and their own objectives, and can only interact via natural language they generate. Each agent, therefore, needs to learn to operate optimally in an environment with multiple sources of uncertainty (its own NLU and NLG, the other agent's NLU, Policy, and NLG). In our evaluation, we show that the stochastic-game agents outperform deep learning based supervised baselines.