ZS4IE: A toolkit for Zero-Shot Information Extraction with simple Verbalizations
This reduces the time and annotation burden for IE analysts, though it is incremental as it builds on existing textual entailment models.
The authors tackled the problem of Information Extraction (IE) requiring annotated training data by introducing a zero-shot workflow where users verbalize entities/relations, achieving very good performance with only 5-15 minutes of user effort per type.
The current workflow for Information Extraction (IE) analysts involves the definition of the entities/relations of interest and a training corpus with annotated examples. In this demonstration we introduce a new workflow where the analyst directly verbalizes the entities/relations, which are then used by a Textual Entailment model to perform zero-shot IE. We present the design and implementation of a toolkit with a user interface, as well as experiments on four IE tasks that show that the system achieves very good performance at zero-shot learning using only 5--15 minutes per type of a user's effort. Our demonstration system is open-sourced at https://github.com/BBN-E/ZS4IE . A demonstration video is available at https://vimeo.com/676138340 .