Metaphor-Extraction-With-GPT-3

KO | EN

gitlite — search

#javascript #hacktoberfest #react #golang #android #machine-learning #deep-learning #linux #docker #awesome #awesome-list #security #ios #mcp

Metaphor-Extraction-With-GPT-3

(★ 12)

Code for our ACL'23 paper on how to identify metaphor mappings with the help of GPT-3

Metaphor-Extraction-With-GPT-3 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# Metaphor Extraction With GPT-3

## Reference

[Paper (PDF)](https://aclanthology.org/2023.acl-long.58/)

[Video Presentation](https://youtu.be/TaReifZ5gy4)

[Related Research](https://lwachowiak.github.io/project/cognitivelinguistics/)

```
@inproceedings{wachowiak2023metaphor,  
  title={{Does GPT-3 Grasp Metaphors? Identifying Metaphor Mappings with Generative Language Models}},   
  author={Wachowiak, Lennart and Gromann, Dagmar},  
  booktitle={Proceedings of the 61th Annual Meeting of the Association for Computational Linguistics (Long Papers)},  
  year={2023}
}
```

## Abstract
Conceptual metaphors present a powerful cognitive vehicle to transfer knowledge structures from a source to a target domain. Prior neural approaches focus on detecting whether natural language sequences are metaphoric or literal. We believe that to truly probe metaphoric knowledge in pre-trained language models, their capability to detect this transfer should be investigated. To this end, this paper proposes to probe the ability of GPT-3 to detect metaphoric language and predict the metaphor’s source do main without any pre-set domains. We experiment with different training sample configurations for fine-tuning and few-shot prompting on two distinct datasets. When provided 12 few-shot samples in the prompt, GPT-3 generates the correct source domain for a new sample with an accuracy of 65.15% in English and 34.65% in Spanish. GPT’s most common error is a hallucinated source domain for which no indicator is present in the sentence. Other common errors include identifying a sequence as literal even though a metaphor is present and predicting the wrong source domain based on specific words in the sequence that are not metaphorically related to the target domain.

## How to Use

- The code for repeating the experiments can be found in the notebook [analysis.ipynb](https://github.com/lwachowiak/Metaphor-Extraction-With-GPT-3/blob/main/analysis.ipynb). The used Python packages are described in the [requirements file](https://github.com/lwachowiak/Metaphor-Extraction-With-GPT-3/blob/main/requirements.txt). An OpenAI API Key is required. 
- Data for prompting and fine-tuning can be found in the [Data](https://github.com/lwachowiak/Metaphor-Extraction-With-GPT-3/tree/main/Data) folder. The LCC corpus should be requested from the authors of the LCC corpus. 
- Annotations made through GPT-3 can be found in the respective folders for [validation](https://github.com/lwachowiak/Metaphor-Extraction-With-GPT-3/tree/main/Validation%20Results/Source%20Completion) and [test](https://github.com/lwachowiak/Metaphor-Extraction-With-GPT-3/tree/main/Test%20Results/Source%20Completion/Few%20Shot) results. The test data also contains the authors' annotations.

![Image](https://raw.githubusercontent.com/lwachowiak/Metaphor-Extraction-With-GPT-3/main/Poster-ACL.png)

Metaphor-Extraction-With-GPT-3

Metaphor-Extraction-With-GPT-3 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Metaphor-Extraction-With-GPT-3

Metaphor-Extraction-With-GPT-3 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README