Zero-Shot Information Extraction as a Unified Text-to-Triple Translation

We cast a suite of information extraction tasks into a text-to-triple translation framework. Instead of solving each task relying on task-specific datasets and models, we formalize the task as a translation between task-specific input text and output triples. By taking the task-specific input, we enable a task-agnostic translation by leveraging the latent knowledge that a pre-trained language model has about the task. We further demonstrate that a simple pre-training task of predicting which relational information corresponds to which input text is an effective way to produce task-specific outputs. This enables the zero-shot transfer of our framework to downstream tasks. We study the zero-shot performance of this framework on open information extraction (OIE2016, NYT, WEB, PENN), relation classification (FewRel and TACRED), and factual probe (Google-RE and T-REx). The model transfers non-trivially to most tasks and is often competitive with a fully supervised method without the need for any task-specific training. For instance, we significantly outperform the F1 score of the supervised open information extraction without needing to use its training set.

PDF Abstract EMNLP 2021 PDF EMNLP 2021 Abstract

Results from the Paper

 Ranked #1 on Open Information Extraction on OIE2016 (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
Relation Classification FewRel DeepEx (zero-shot top-1) F1 48.8 # 1
Relation Classification FewRel DeepEx (zero-shot top-10) F1 92.9 # 2
Open Information Extraction NYT DeepEx (zero-shot) F1 85.5 # 1
AUC 72.5 # 1
Open Information Extraction OIE2016 DeepEx (zero-shot) F1 72.6 # 1
AUC 58.6 # 1
Open Information Extraction Penn Treebank DeepEx (zero-shot) F1 88.5 # 1
AUC 81.5 # 1
Relation Classification TACRED DeepEx (zero-shot top-10) F1 76.4 # 4
Relation Classification TACRED DeepEx (zero-shot top-1) F1 49.2 # 1
Open Information Extraction Web DeepEx (zero-shot) F1 91.2 # 1
AUC 82.4 # 1


No methods listed for this paper. Add relevant methods here