ExT5: Towards Extreme Multi-Task Scaling for Transfer Learning

Vamsi Aribandi · Yi Tay · Tal Schuster · Jinfeng Rao · Huaixiu Steven Zheng · Sanket Vaibhav Mehta · Honglei Zhuang · Vinh Tran · Dara Bahri · Jianmo Ni · Jai Gupta · Kai Hui · Sebastian Ruder · Donald Metzler

Keywords: [ natural language processing ] [ multi-task learning ] [ transfer learning ]

[ Abstract ]
[ Visit Poster at Spot E3 in Virtual World ] [ OpenReview
Wed 27 Apr 10:30 a.m. PDT — 12:30 p.m. PDT


Despite the recent success of multi-task learning and transfer learning for natural language processing (NLP), few works have systematically studied the effect of scaling up the number of tasks during pre-training. Towards this goal, this paper introduces ExMix (Extreme Mixture): a massive collection of 107 supervised NLP tasks across diverse domains and task-families. Using ExMix, we study the effect of multi-task pre-training at the largest scale to date, and analyze co-training transfer amongst common families of tasks. Through this analysis, we show that manually curating an ideal set of tasks for multi-task pre-training is not straightforward, and that multi-task scaling can vastly improve models on its own. Finally, we propose ExT5: a model pre-trained using a multi-task objective of self-supervised span denoising and supervised ExMix. Via extensive experiments, we show that ExT5 outperforms strong T5 baselines on SuperGLUE, GEM, Rainbow, Closed-Book QA tasks, and several tasks outside of ExMix. ExT5 also significantly improves sample efficiency while pre-training.

Chat is not available.