# Learning through Dialogue Interactions by Asking Questions

arXiv: https://arxiv.org/abs/1612.04936 · HTML: https://mlsanity.com/paper/1612.04936/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 15 Dec 2016.

## Built on by (most cited first, 5 of 5)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [BlenderBot 3: a deployed conversational agent that continually learns to responsibly engage](https://mlsanity.com/paper/2208.03188/) (2022)
- [Recent Advances in Deep Learning Based Dialogue Systems: A Systematic Survey](https://mlsanity.com/paper/2105.04387/) (2021)
- [Studying Catastrophic Forgetting in Neural Ranking Models](https://mlsanity.com/paper/2101.06984/) (2021)
- [Ericson: An Interactive Open-Domain Conversational Search Agent](https://mlsanity.com/paper/2304.02233/) (2023)

## Abstract (arXiv)

A good dialogue agent should have the ability to interact with users by both responding to questions and by asking questions, and importantly to learn from both types of interaction. In this work, we explore this direction by designing a simulator and a set of synthetic tasks in the movie domain that allow such interactions between a learner and a teacher. We investigate how a learner can benefit from asking questions in both offline and online reinforcement learning settings, and demonstrate that the learner improves when asking questions. Finally, real experiments with Mechanical Turk validate the approach. Our work represents a first step in developing such end-to-end learned interactive dialogue agents.