Summary of Synergetic Event Understanding: a Collaborative Approach to Cross-document Event Coreference Resolution with Large Language Models, by Qingkai Min et al.
Synergetic Event Understanding: A Collaborative Approach to Cross-Document Event Coreference Resolution with Large Language Models
by Qingkai Min, Qipeng Guo, Xiangkun Hu, Songfang Huang, Zheng Zhang, Yue Zhang
First submitted to arxiv on: 4 Jun 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed collaborative approach for cross-document event coreference resolution (CDECR) leverages the capabilities of both large language models (LLMs) like ChatGPT and small language models (SLMs) like BERT. The method begins with the LLM summarizing events through prompting, which is then used to refine the SLM’s learning of event representations during fine-tuning. Experimental results demonstrate that this approach surpasses the performance of both individual models, achieving state-of-the-art performance across various datasets. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Cross-document event coreference resolution helps computers understand when different documents are talking about the same real-world events. The problem is hard because documents have different contexts and ways of expressing ideas. Researchers used to solve this problem by fine-tuning small language models like BERT, but these models can learn simple patterns rather than understanding complex meanings. New large language models like ChatGPT are better at understanding context, but they struggle to adapt to specific tasks. This paper proposes a new approach that combines the strengths of both types of models. The result is a system that performs better than either individual model alone. |
Keywords
» Artificial intelligence » Bert » Coreference » Fine tuning » Prompting
