Papers by Biaoshuai Zheng
EventRelBench: A Comprehensive Benchmark for Evaluating Event Relation Understanding in Large Language Models (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing LLMs fail to capture event relationships, despite advances in NLP . a new benchmark is being developed to assess LLM's ability to extract event relationships . |
| Approach: | They propose a benchmark to assess LLMs' ability to extract event relations . EventRelBench comprises 35K diverse event relation questions . |
| Outcome: | The benchmark EventRelBench measures the performance of large language models on event relation extraction tasks. |