Effective Distant Supervision for Temporal Relation Extraction
Published in
(2020)External Links:
TL
TL;DR
It is demonstrated that a pre-trained Transformer model is able to transfer from the weakly labeled examples to human-annotated benchmarks in both zero-shot and few-shot settings, and that the masking scheme is important in improving generalization.
Abstract
A principal barrier to training temporal relation extraction models in new domains is the lack of varied, high quality examples and the challenge of collecting more. We present a method of automatically collecting distantly-supervised examples of temporal relations. We scrape and automatically label event pairs where the temporal relations are made explicit in text, then mask out those explicit cues, forcing a model trained on this data to learn other signals. We demonstrate that a pre-trained Transformer model is able to transfer from the weakly labeled examples to human-annotated benchmarks in both zero-shot and few-shot settings, and that the masking scheme is important in improving generalization.
References30 items
1
Conditional Generation of Temporally-ordered Event Sequences
2
AutoQA: From Databases to Q&A Semantic Parsers with Only Synthetic Training Data
3
Temporal Common Sense Acquisition with Minimal Supervision
4
Unnatural Language Processing: Bridging the Gap Between Synthetic and Natural Language Data
5