jihyoung/rebot-generation

👫ReBot - Generation Module⏰

ReBot is a novel multi-session dialgoue model which can generate dialogue with chronological dynamics! ReBot consists two modules: (1) chronological summarization module; (2) dialogue generation module.

This repoistory for dialogue generation module. You can check summarization module on this repoistory.

🚨 Please be cautious when testing our model with the Hosted Inference API. Our model takes sequences as input, so you should provide sequences as input through the API as well.

Model description

• Paper: Conversation Chronicles: Towards Diverse Temporal and Relational Dynamics in Multi-Session Conversations
• Dataset : Conversation Chronicles
• Generation Module of Model : this repoistory
• Summarization Module of Model : chronological summarization module

Load with Transformers

To load our dataset with Hugging Face Transformers, please use the following code:

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

tokenizer = AutoTokenizer.from_pretrained("jihyoung/rebot-generation")
model = AutoModelForSeq2SeqLM.from_pretrained("jihyoung/rebot-generation")

Citation Information

@inproceedings{jang-etal-2023-conversation,
    title = "Conversation Chronicles: Towards Diverse Temporal and Relational Dynamics in Multi-Session Conversations",
    author = "Jang, Jihyoung  and
      Boo, Minseong  and
      Kim, Hyounghun",
    editor = "Bouamor, Houda  and
      Pino, Juan  and
      Bali, Kalika",
    booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
    month = dec,
    year = "2023",
    address = "Singapore",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2023.emnlp-main.838",
    doi = "10.18653/v1/2023.emnlp-main.838",
    pages = "13584--13606",
}