Fabrice Harel-Canada — accepted AI-conference papers

2022

EnDex: Evaluation of Dialogue Engagingness at Scale

Guangxuan Xu, Ruibo Liu, Fabrice Harel-Canada, Nischal Reddy Chandra, Nanyun Peng

EMNLP 2022finding

We propose EnDex, the first human-reaction based model to evaluate dialogue engagingness. EnDex is trained on 80k Reddit-based Engagement Dataset (RED) curated using a novel distant-supervision framework. Engagingness is a key measure that captures high-level quality of AI dialogue systems and close…

Cited by 12 Source ↗PDF ↗code ↗Scholar ↗

2022

Sibylvariant Transformations for Robust Text Classification

Fabrice Harel-Canada, Muhammad Ali Gulzar, Nanyun Peng, Miryung Kim

ACL 2022findings

The vast majority of text transformation techniques in NLP are inherently limited in their ability to expand input space coverage due to an implicit constraint to preserve the original class label. In this work, we propose the notion of sibylvariance (SIB) to describe the broader set of transforms t…

Cited by 1 Source ↗PDF ↗code ↗Scholar ↗