MVP-story

The MVP-story model was proposed in MVP: Multi-task Supervised Pre-training for Natural Language Generation by Tianyi Tang, Junyi Li, Wayne Xin Zhao and Ji-Rong Wen.

The detailed information and instructions can be found https://github.com/RUCAIBox/MVP.

Model Description

MVP-story is a prompt-based model that MVP is further equipped with prompts pre-trained using labeled story generation datasets. It is a variant (MVP+S) of our main MVP model. It follows a Transformer encoder-decoder architecture with layer-wise prompts.

MVP-story is specially designed for story generation tasks, such as ROCStories and WritingPrompts.

Example

>>> from transformers import MvpTokenizer, MvpForConditionalGeneration

>>> tokenizer = MvpTokenizer.from_pretrained("RUCAIBox/mvp")
>>> model = MvpForConditionalGeneration.from_pretrained("RUCAIBox/mvp-story")

>>> inputs = tokenizer(
...     "Given the story title: I think all public schools should have a uniform dress code.",
...     return_tensors="pt",
... )
>>> generated_ids = model.generate(**inputs, max_length=1024)
>>> tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
['I think it would be a good idea to have uniform dress codes for all public schools. It would make it easier for students to dress appropriately.']

Related Models

MVP: https://huggingface.co/RUCAIBox/mvp.

Prompt-based models:

Multi-task models:

Citation

@article{tang2022mvp,
  title={MVP: Multi-task Supervised Pre-training for Natural Language Generation},
  author={Tang, Tianyi and Li, Junyi and Zhao, Wayne Xin and Wen, Ji-Rong},
  journal={arXiv preprint arXiv:2206.12131},
  year={2022},
  url={https://arxiv.org/abs/2206.12131},
}
Downloads last month
132
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.