LLaMA-ABSA

KO | EN

gitlite — search

#javascript #ai #golang #rust #deep-learning #linux #nodejs #macos #kubernetes #vue #pytorch #mcp #ios #open-source

(★ 10)

No description available.

LLaMA-ABSA 최신버젼 다운로드

최종 버전 다운로드 (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

# LLaMA-ABSA

## Introduction
This repository contains the code for the paper "LLaMA-Based Models for Aspect-Based Sentiment Analysis".

## Requirements
- Python 3.10
- Required packages are listed in `requirements.txt`

## Datasets
- Data is available in the `data_sota` folder.

## Running the code
- To run the code for instruction tuning, execute the following command:
```
python main.py --model microsoft/Orca-2-13b --data_path tasd/rest16 --instruction_tuning --no_wandb
```

- For few-shot scenario, execute the following command:
```
python main.py --model microsoft/Orca-2-13b --data_path tasd/rest16 --few_shot_prompt --no_wandb
```

- For zero-shot scenario, execute the following command:
```
python main.py --model microsoft/Orca-2-13b --data_path tasd/rest16 --no_wandb
```

- Model can be changed by changing the `--model` argument.
- Dataset can be changed by changing the `--data_path` argument.
- For more details, refer to the `main.py` and `src/args_utils.py` files.

## Citation
If you find this repository helpful for your research, please cite our paper as follows:
```
@inproceedings{smid-etal-2024-llama,
    title = "{LL}a{MA}-Based Models for Aspect-Based Sentiment Analysis",
    author = "{\v{S}}m{\'i}d, Jakub  and
      Priban, Pavel  and
      Kral, Pavel",
    editor = "De Clercq, Orph{\'e}e  and
      Barriere, Valentin  and
      Barnes, Jeremy  and
      Klinger, Roman  and
      Sedoc, Jo{\~a}o  and
      Tafreshi, Shabnam",
    booktitle = "Proceedings of the 14th Workshop on Computational Approaches to Subjectivity, Sentiment, {\&} Social Media Analysis",
    month = aug,
    year = "2024",
    address = "Bangkok, Thailand",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2024.wassa-1.6/",
    doi = "10.18653/v1/2024.wassa-1.6",
    pages = "63--70",
    abstract = "While large language models (LLMs) show promise for various tasks, their performance in compound aspect-based sentiment analysis (ABSA) tasks lags behind fine-tuned models. However, the potential of LLMs fine-tuned for ABSA remains unexplored. This paper examines the capabilities of open-source LLMs fine-tuned for ABSA, focusing on LLaMA-based models. We evaluate the performance across four tasks and eight English datasets, finding that the fine-tuned Orca 2 model surpasses state-of-the-art results in all tasks. However, all models struggle in zero-shot and few-shot scenarios compared to fully fine-tuned ones. Additionally, we conduct error analysis to identify challenges faced by fine-tuned models."
}
```

LLaMA-ABSA

LLaMA-ABSA 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

LLaMA-ABSA

LLaMA-ABSA 최신버젼 다운로드

오늘의 추천 프로젝트

Was this content helpful?

Recent Feedback

Download README