Document-Level Event Argument Extraction by Conditional Generation

By Sha Li, Heng Ji, Jiawei Han. North American Chapter of the Association for Computational Linguistics, 2021

376 citations recorded by the source corpus.

Event extraction has long been treated as a sentence-level task in the IE community. We argue that this setting does not match human informative seeking behavior and leads to incomplete and uninformative extraction results. We propose a document-level neural event argument extraction model by formulating the task as conditional generation following event templates. We also compile a new document-level event extraction benchmark dataset WikiEvents which includes complete event and coreference annotation. On the task of argument extraction, we achieve an absolute gain of 7.6% F1 and 5.7% F1 over the next best model on the RAMS and WikiEvents dataset respectively. On the more challenging task of informative argument extraction, which requires implicit coreference reasoning, we achieve a 9.3% F1 gain over the best baseline. To demonstrate the portability of our model, we also create the first end-to-end zero-shot event extraction framework and achieve 97% of fully supervised model’s trigger extraction performance and 82% of the argument extraction performance given only access to 10 out of the 33 types on ACE.

Subfields: Artificial Intelligence, Information Systems

View Paper: https://www.semanticscholar.org/paper/2a15f541a8216782a7579f01f56762448e48b582

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Diyi Yang
2. Jiaao Chen
3. Xingyao Wang
4. Sha Li
5. Xianpei Han
6. Heng Ji
7. Hongyu Lin
8. Le Sun

In Information Systems

1. Xingyao Wang
2. Sha Li
3. Heng Ji
4. Caleb Ziems
5. Diyi Yang
6. Le Sun
7. Xianpei Han

Citations in the corpus

• Code4Struct: Code Generation for Few-Shot Structured Prediction from Natural Language (2022, methodology): “BART-Gen (Li et al., 2021) <arg1> transported <arg2> in <arg3> vehicle from <arg4> place to <arg5> place Text2Event (Lu et al., 2021) Chowdhery et al., 2022) trained on web-scale language corpora.”
• Unified Structure Generation for Universal Information Extraction (2022, methodology): “The generation can be enhanced using label templates (Li et al., 2021a; Liu et al., 2021; Cui et al., 2021), schema (Lu et al., 2021; Ahmad et al., 2021), and augmented language methods (Paolini et al., 2021).”
• Large language models for generative information extraction: a survey (2023, neutral): “Furthermore, generative methods significantly outperform discriminative ones in event extraction tasks, particularly in argument classification, highlighting the substantial potential of generative LLMs in this domain.”
• Can Large Language Models Transform Computational Social Science? (2023, background): “WikiEvents (Li et al., 2021) is a documentlevel event extraction benchmark for news articles that were linked from English Wikipedia articles.”