Learning as Conversation: Dialogue Systems Reinforced for Information Acquisition
This addresses the problem of knowledge acquisition for users who prefer conversational learning over traditional reading, though it appears incremental as it builds on existing dialogue and reinforcement learning methods.
The paper tackles the problem of enabling users to acquire knowledge through conversation instead of reading, by developing a dialogue system that uses reinforced self-play to operate across domains without in-domain data. The result is a system that delivers informative and attentive conversations, helping users substantially gain knowledge without reading passages, as demonstrated through evaluations on three large public datasets.
We propose novel AI-empowered chat bots for learning as conversation where a user does not read a passage but gains information and knowledge through conversation with a teacher bot. Our information-acquisition-oriented dialogue system employs a novel adaptation of reinforced self-play so that the system can be transferred to various domains without in-domain dialogue data, and can carry out conversations both informative and attentive to users. Our extensive subjective and objective evaluations on three large public data corpora demonstrate the effectiveness of our system to deliver knowledge-intensive and attentive conversations and help end users substantially gain knowledge without reading passages. Our code and datasets are publicly available for follow-up research.