# Improving Generalization in Coreference Resolution via Adversarial Training

arXiv: https://arxiv.org/abs/1908.04728 · HTML: https://mlsanity.com/paper/1908.04728/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 13 Aug 2019.

## Built on by (most cited first, 10 of 10)

- [Entity-Based Knowledge Conflicts in Question Answering](https://mlsanity.com/paper/2109.05052/) (2021)
- [State-of-the-art generalisation research in NLP: A taxonomy and review](https://mlsanity.com/paper/2210.03050/) (2022)
- [On Generalization in Coreference Resolution](https://mlsanity.com/paper/2109.09667/) (2021)
- [On the Robustness of Reading Comprehension Models to Entity Renaming](https://mlsanity.com/paper/2110.08555/) (2021)
- [Challenges to Evaluating the Generalization of Coreference Resolution Models: A Measurement Modeling Perspective](https://mlsanity.com/paper/2303.09092/) (2023)
- [Concept Identification of Directly and Indirectly Related Mentions Referring to Groups of Persons](https://mlsanity.com/paper/2107.00955/) (2021)
- [Sentence-Incremental Neural Coreference Resolution](https://mlsanity.com/paper/2305.16947/) (2023)
- [Incorporating Singletons and Mention-based Features in Coreference Resolution via Multi-task Learning for Better Generalization](https://mlsanity.com/paper/2309.11582/) (2023)
- [Review of coreference resolution in English and Persian](https://mlsanity.com/paper/2211.04428/) (2022)
- [Efficient Seq2seq Coreference Resolution Using Entity Representations](https://mlsanity.com/paper/2510.14504/) (2025)

## Abstract (arXiv)

In order for coreference resolution systems to be useful in practice, they must be able to generalize to new text. In this work, we demonstrate that the performance of the state-of-the-art system decreases when the names of PER and GPE named entities in the CoNLL dataset are changed to names that do not occur in the training set. We use the technique of adversarial gradient-based training to retrain the state-of-the-art system and demonstrate that the retrained system achieves higher performance on the CoNLL dataset (both with and without the change of named entities) and the GAP dataset.