PRESTO: A Multilingual Dataset for Parsing Realistic Task-Oriented Dialogs

Rahul Goel, Waleed Ammar, Aditya Gupta, Siddharth Vashishtha, Motoki Sano +11 more
3/15/2023
cs.CL

No official code found. Best implementation: google-research-datasets/presto (116 stars, last commit 2023-03-17).

Abstract

Research interest in task-oriented dialogs has increased as systems such as Google Assistant, Alexa and Siri have become ubiquitous in everyday life. However, the impact of academic research in this area has been limited by the lack of datasets that realistically capture the wide array of user pain points. To enable research on some of the more challenging aspects of parsing realistic conversations, we introduce PRESTO, a public dataset of over 550K contextual multilingual conversations between humans and virtual assistants. PRESTO contains a diverse array of challenges that occur in real-world NLU tasks such as disfluencies, code-switching, and revisions. It is the only large scale human generated conversational parsing dataset that provides structured context such as a user's contacts and lists for each example. Our mT5 model based baselines demonstrate that the conversational phenomenon present in PRESTO are challenging to model, which is further pronounced in a low-resource setup.

View on arXivView PDF

Code Implementations(1)

google-research-datasets/prestoCommunity100%

A Multilingual Dataset for Parsing Realistic Task-Oriented Dialogs

1166Jan 31, 20233 years ago

Similar Papers

SpokenUS: A Spoken User Simulator for Task-Oriented Dialogue

86% match2026

Multi3WOZ: A Multilingual, Multi-Domain, Multi-Parallel Dataset for Training and Evaluating Culturally Adapted Task-Oriented Dialog Systems

86% match1 repo172023

DialogStudio: Towards Richest and Most Diverse Unified Dataset Collection for Conversational AI

86% match1 repo5272023

PLACES: Prompting Language Models for Social Conversation Synthesis

85% match1 repo112023

Does Collaborative Human-LM Dialogue Generation Help Information Extraction from Human Dialogues?

85% match2023

SalesBot 2.0: A Human-Like Intent-Guided Chit-Chat Dataset

85% match2023

Cite this paper

@article{goel2023presto,
  title  = {PRESTO: A Multilingual Dataset for Parsing Realistic Task-Oriented Dialogs},
  author = {Rahul Goel and Waleed Ammar and Aditya Gupta and Siddharth Vashishtha and Motoki Sano and Faiz Surani and Max Chang and HyunJeong Choe and David Greene and Kyle He and Rattima Nitisaroj and Anna Trukhina and Shachi Paul and Pararth Shah and Rushin Shah and Zhou Yu},
  year   = {2023},
  eprint = {2303.08954},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2303.08954v2}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2303.08954.svg)](https://codeofpaper.com/paper/2303.08954)

Discussion