LLM4ED

The implementation and data of the paper: Harnessing the Power of Large Language Models for Empathetic Response Generation: Empirical Investigations and Improvement.

Requirements

• torch
• nltk
• transformers
• spacy
• sentence-transformers

Preparation

• Download the COMET-BART checkpoint and place it in data/Comet.
• You should have OpenAI api-key.

Usage

bash main.sh

Proposal

The data from this work is available here.
You can leverage the data generated by powerful LLMs (e.g., ChatGPT) to empower other LLMs to improve their empathy capabilities.

Citation

If our work is useful for your research, please kindly cite our paper as follows:

@article{qian2023harnessing,
      title={Harnessing the Power of Large Language Models for Empathetic Response Generation: Empirical Investigations and Improvements},
      author={Qian, Yushan and Zhang, Wei-Nan and Liu, Ting},
      journal={arXiv preprint arXiv:2310.05140},
      year={2023},
}