# Open Relation and Event Type Discovery with Type Abstraction

arXiv: https://arxiv.org/abs/2212.00178 · HTML: https://mlsanity.com/paper/2212.00178/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 19 of its 51 references that are in the map.
- First posted on arXiv 30 Nov 2022.

## Builds on (most cited first, 19 of 19)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [SimCSE: Simple Contrastive Learning of Sentence Embeddings](https://mlsanity.com/paper/2104.08821/) (2021)
- [Matching the Blanks: Distributional Similarity for Relation Learning](https://mlsanity.com/paper/1906.03158/) (2019)
- [FewRel: A Large-Scale Supervised Few-Shot Relation Classification Dataset with State-of-the-Art Evaluation](https://mlsanity.com/paper/1810.10147/) (2018)
- [Event Extraction by Answering (Almost) Natural Questions](https://mlsanity.com/paper/2004.13625/) (2020)
- [Document-Level Event Argument Extraction by Conditional Generation](https://mlsanity.com/paper/2104.05919/) (2021)
- [Learning to cluster in order to transfer across domains and tasks](https://mlsanity.com/paper/1711.10125/) (2017)
- [Packed Levitated Marker for Entity and Relation Extraction](https://mlsanity.com/paper/2109.06067/) (2021)
- [SelfORE: Self-supervised Relational Feature Learning for Open Relation Extraction](https://mlsanity.com/paper/2004.02438/) (2020)
- [A Frustratingly Easy Approach for Entity and Relation Extraction](https://mlsanity.com/paper/2010.12812/) (2020)
- [Hearst Patterns Revisited: Automatic Hypernym Detection from Large Text Corpora](https://mlsanity.com/paper/1806.03191/) (2018)
- [Every child should have parents: a taxonomy refinement algorithm based on hyperbolic term embeddings](https://mlsanity.com/paper/1906.02002/) (2019)
- [Constructing Taxonomies from Pretrained Language Models](https://mlsanity.com/paper/2010.12813/) (2020)
- [CoRel: Seed-Guided Topical Taxonomy Construction by Concept Learning and Relation Transferring](https://mlsanity.com/paper/2010.06714/) (2020)
- [Corpus-based Open-Domain Event Type Induction](https://mlsanity.com/paper/2109.03322/) (2021)
- [A Relation-Oriented Clustering Method for Open Relation Extraction](https://mlsanity.com/paper/2109.07205/) (2021)
- [Revisiting Unsupervised Relation Extraction](https://mlsanity.com/paper/2005.00087/) (2020)
- [Element Intervention for Open Relation Extraction](https://mlsanity.com/paper/2106.09558/) (2021)
- [SynSetExpan: An Iterative Framework for Joint Entity Set Expansion and Synonym Discovery](https://mlsanity.com/paper/2009.13827/) (2020)

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

- [ClusterLLM: Large Language Models as a Guide for Text Clustering](https://mlsanity.com/paper/2305.14871/) (2023)
- [Language Models Can Improve Event Prediction by Few-Shot Abductive Reasoning](https://mlsanity.com/paper/2305.16646/) (2023)
- [Grasping the Essentials: Tailoring Large Language Models for Zero-Shot Relation Extraction](https://mlsanity.com/paper/2402.11142/) (2024)
- [Event Detection from Social Media for Epidemic Prediction](https://mlsanity.com/paper/2404.01679/) (2024)
- [Harvesting Event Schemas from Large Language Models](https://mlsanity.com/paper/2305.07280/) (2023)
- [Category Discovery: An Open-World Perspective](https://mlsanity.com/paper/2509.22542/) (2025)
- [Learning from Semi-Factuals: A Debiased and Semantic-Aware Framework for Generalized Relation Discovery](https://mlsanity.com/paper/2401.06327/) (2024)
- [Open-world Semi-supervised Generalized Relation Discovery Aligned in a Real-world Setting](https://mlsanity.com/paper/2305.13533/) (2023)
- [Towards a More Generalized Approach in Open Relation Extraction](https://mlsanity.com/paper/2505.22801/) (2025)
- [CLUE: A Contrastive Learning Framework for Unsupervised Event Type Induction and Annotation](https://mlsanity.com/paper/2610.12580/) (2026)

## Abstract (arXiv)

Conventional closed-world information extraction (IE) approaches rely on human ontologies to define the scope for extraction. As a result, such approaches fall short when applied to new domains. This calls for systems that can automatically infer new types from given corpora, a task which we refer to as type discovery. To tackle this problem, we introduce the idea of type abstraction, where the model is prompted to generalize and name the type. Then we use the similarity between inferred names to induce clusters. Observing that this abstraction-based representation is often complementary to the entity/trigger token representation, we set up these two representations as two views and design our model as a co-training framework. Our experiments on multiple relation extraction and event extraction datasets consistently show the advantage of our type abstraction approach. Code available at https://github.com/raspberryice/type-discovery-abs.