Beyond Pairwise: Global Zero-shot Temporal Graph Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914047950061568 |
|---|---|
| author | Eirew, Alon Bar, Kfir Dagan, Ido |
| author_facet | Eirew, Alon Bar, Kfir Dagan, Ido |
| contents | Temporal relation extraction (TRE) is a fundamental task in natural language processing (NLP) that involves identifying the temporal relationships between events in a document. Despite the advances in large language models (LLMs), their application to TRE remains limited. Most existing approaches rely on pairwise classification, where event pairs are classified in isolation, leading to computational inefficiency and a lack of global consistency in the resulting temporal graph. In this work, we propose a novel zero-shot method for TRE that generates a document's complete temporal graph in a single step, followed by temporal constraint optimization to refine predictions and enforce temporal consistency across relations. Additionally, we introduce OmniTemp, a new dataset with complete annotations for all pairs of targeted events within a document. Through experiments and analyses, we demonstrate that our method outperforms existing zero-shot approaches and offers a competitive alternative to supervised TRE models. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_11114 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Beyond Pairwise: Global Zero-shot Temporal Graph Generation Eirew, Alon Bar, Kfir Dagan, Ido Computation and Language Temporal relation extraction (TRE) is a fundamental task in natural language processing (NLP) that involves identifying the temporal relationships between events in a document. Despite the advances in large language models (LLMs), their application to TRE remains limited. Most existing approaches rely on pairwise classification, where event pairs are classified in isolation, leading to computational inefficiency and a lack of global consistency in the resulting temporal graph. In this work, we propose a novel zero-shot method for TRE that generates a document's complete temporal graph in a single step, followed by temporal constraint optimization to refine predictions and enforce temporal consistency across relations. Additionally, we introduce OmniTemp, a new dataset with complete annotations for all pairs of targeted events within a document. Through experiments and analyses, we demonstrate that our method outperforms existing zero-shot approaches and offers a competitive alternative to supervised TRE models. |
| title | Beyond Pairwise: Global Zero-shot Temporal Graph Generation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2502.11114 |