Zero-Shot Generalizable End-to-End Task-Oriented Dialog System using Context Summarization and Domain Schema

Resumo

Task-oriented dialog systems empower users to accom-
plish their goals by facilitating intuitive and expres-
sive natural language interactions. State-of-the-art ap-
proaches in task-oriented dialog systems formulate the
problem as a conditional sequence generation task and
fine-tune pre-trained causal language models in the su-
pervised setting. This requires labeled training data
for each new domain or task, and acquiring such data
is prohibitively laborious and expensive, thus making
it a bottleneck for scaling systems to a wide range
of domains. To overcome this challenge, we intro-
duce a novel Zero-Shot generalizable end-to-end Task-
oriented Dialog system, ZS-ToD, that leverages domain
schemas to allow for robust generalization to unseen do-
mains and exploits effective summarization of the dia-
log history. We employ GPT-2 as a backbone model and
introduce a two-step training process where the goal of
the first step is to learn the general structure of the dialog
data and the second step optimizes the response gen-
eration as well as intermediate outputs, such as dialog
state and system actions. As opposed to state-of-the-art
systems that are trained to fulfill certain intents in the
given domains and memorize task-specific conversa-
tional patterns, ZS-ToD learns generic task-completion
skills by comprehending domain semantics via domain
schemas and generalizing to unseen domains seam-
lessly. We conduct an extensive experimental evaluation
on SGD and SGD-X datasets that span up to 20 unique
domains and ZS-ToD outperforms state-of-the-art sys-
tems on key metrics, with an improvement of +17% on
joint goal accuracy and +5 on inform. Additionally,
we present a detailed ablation study to demonstrate the
effectiveness of the proposed components and training
mechanism.