Zero-Shot Dialogue Disentanglement by Self-Supervised Entangled Response Selection
This addresses the quadratic labeling effort for dialogue disentanglement, useful for discourse analysis and downstream applications, though it is incremental as it builds on existing response selection methods.
The paper tackles the problem of dialogue disentanglement, which groups utterances into threads, by proposing a zero-shot solution that achieves a cluster F1 score of 25 without labeled data, and with only 10% of labeled data, it nearly matches full dataset performance.
Dialogue disentanglement aims to group utterances in a long and multi-participant dialogue into threads. This is useful for discourse analysis and downstream applications such as dialogue response selection, where it can be the first step to construct a clean context/response set. Unfortunately, labeling all~\emph{reply-to} links takes quadratic effort w.r.t the number of utterances: an annotator must check all preceding utterances to identify the one to which the current utterance is a reply. In this paper, we are the first to propose a~\textbf{zero-shot} dialogue disentanglement solution. Firstly, we train a model on a multi-participant response selection dataset harvested from the web which is not annotated; we then apply the trained model to perform zero-shot dialogue disentanglement. Without any labeled data, our model can achieve a cluster F1 score of 25. We also fine-tune the model using various amounts of labeled data. Experiments show that with only 10\% of the data, we achieve nearly the same performance of using the full dataset\footnote{Code is released at \url{https://github.com/chijames/zero_shot_dialogue_disentanglement}}.