SIFiD: Reassess Summary Factual Inconsistency Detection with LLM
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866911793659510784 |
|---|---|
| author | Yang, Jiuding Liu, Hui Guo, Weidong Rao, Zhuwei Xu, Yu Niu, Di |
| author_facet | Yang, Jiuding Liu, Hui Guo, Weidong Rao, Zhuwei Xu, Yu Niu, Di |
| contents | Ensuring factual consistency between the summary and the original document is paramount in summarization tasks. Consequently, considerable effort has been dedicated to detecting inconsistencies. With the advent of Large Language Models (LLMs), recent studies have begun to leverage their advanced language understanding capabilities for inconsistency detection. However, early attempts have shown that LLMs underperform traditional models due to their limited ability to follow instructions and the absence of an effective detection methodology. In this study, we reassess summary inconsistency detection with LLMs, comparing the performances of GPT-3.5 and GPT-4. To advance research in LLM-based inconsistency detection, we propose SIFiD (Summary Inconsistency Detection with Filtered Document) that identify key sentences within documents by either employing natural language inference or measuring semantic similarity between summaries and documents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_07557 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SIFiD: Reassess Summary Factual Inconsistency Detection with LLM Yang, Jiuding Liu, Hui Guo, Weidong Rao, Zhuwei Xu, Yu Niu, Di Computation and Language Machine Learning Ensuring factual consistency between the summary and the original document is paramount in summarization tasks. Consequently, considerable effort has been dedicated to detecting inconsistencies. With the advent of Large Language Models (LLMs), recent studies have begun to leverage their advanced language understanding capabilities for inconsistency detection. However, early attempts have shown that LLMs underperform traditional models due to their limited ability to follow instructions and the absence of an effective detection methodology. In this study, we reassess summary inconsistency detection with LLMs, comparing the performances of GPT-3.5 and GPT-4. To advance research in LLM-based inconsistency detection, we propose SIFiD (Summary Inconsistency Detection with Filtered Document) that identify key sentences within documents by either employing natural language inference or measuring semantic similarity between summaries and documents. |
| title | SIFiD: Reassess Summary Factual Inconsistency Detection with LLM |
| topic | Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2403.07557 |