Welcome to Sólo Escúchame – an open-source Spanish emotional assistance chatbot powered by the LLaMA-2-7b-Chat model. Our goal is to provide emotional support through advanced AI. 💡
In this repository, you'll find:
-
Fine-Tuning Scripts: Explore the scripts we use to fine-tune the LLaMA-2-7b-Chat model with our custom datasets.
-
Datasets: Access links to our curated datasets.
-
Research Paper: Read our detailed research paper to understand the methodology and findings.
-
Additional Details: Discover other relevant information and updates.
Sólo Escúchame is an enhanced, fine-tuned model based on LLaMA-2-7b-Chat, meticulously trained using the datasets outlined in the next section. You can download the model and find detailed specifications by clicking here.
Additionally, we offer another version of this model, also fine-tuned on LLaMA-2-7b-Chat, utilizing the PEFT Adapter. This version is optimized for broader accessibility, requiring fewer computational resources. For more details and to download this model, click here.
At the core of our work lies the data. We have introduced two essential datasets:
- Hispanic Emotion Recognition Based on Plutchik’s Wheel (HRECPW) - is designed for training models in the task of emotional classification. This dataset allows models to identify and classify various emotional categories based on the Plutchik wheel, with a particular focus on the Spanish language.
- HEAR (Hispanic Emotional Accompaniment Responses) - is designed to train language models in the task of emotionally accompanying users. This dataset enables models to generate empathetic and appropriate responses in Spanish, understanding and responding to different emotional situations.
Our paper has been accepted in the at the 23rd Mexican International Conference on Artificial Intelligence (MICAI 2024). You can also find our work on arxiv.
If you use this code or find our work helpful, please cite our paper:
@misc{Gil2024, title={S\'olo Esc\'uchame: Spanish Emotional Accompaniment Chatbot}, author={Bruno Gil Ramírez and Jessica López Espejel and María del Carmen Santiago Díaz and Gustavo Trinidad Rubín Linares}, year={2024}, eprint={2408.01852}, archivePrefix={arXiv}, primaryClass={cs.CL}, url={https://arxiv.org/abs/2408.01852}, }
We are working on the chatbot interface ...