---
license: apache-2.0
language:
- en
tags:
- text-generation
- text2text-generation
- summarization
- conversational
pipeline_tag: text-generation

---

# MVP
The MVP model was proposed in [**MVP: Multi-task Supervised Pre-training for Natural Language Generation**](https://github.com/RUCAIBox/MVP/blob/main/paper.pdf) 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](https://github.com/RUCAIBox/MVP).

## Model Description
MVP is supervised pre-trained using a mixture of labeled datasets. It follows a standard Transformer encoder-decoder architecture.

MVP is specially designed for natural language generation and can be adapted to a wide range of generation tasks, including but not limited to summarization, data-to-text generation, open-ended dialogue system, story generation, question answering, question generation, task-oriented dialogue system, commonsense generation, paraphrase generation, text style transfer, and text simplification. Our model can also be adapted to natural language understanding tasks such as sequence classification and (extractive) question answering.

## Examples
For summarization:
```python
>>> from transformers import MvpTokenizer, MvpForConditionalGeneration

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

>>> inputs = tokenizer(
...     "Summarize: You may want to stick it to your boss and leave your job, but don't do it if these are your reasons.",
...     return_tensors="pt",
... )
>>> generated_ids = model.generate(**inputs)
>>> tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
["Why You Shouldn't Quit Your Job"]
```

For data-to-text generation:
```python
>>> from transformers import MvpTokenizerFast, MvpForConditionalGeneration

>>> tokenizer = MvpTokenizerFast.from_pretrained("RUCAIBox/mvp")
>>> model = MvpForConditionalGeneration.from_pretrained("RUCAIBox/mvp")

>>> inputs = tokenizer(
...     "Describe the following data: Iron Man | instance of | Superhero [SEP] Stan Lee | creator | Iron Man",
...     return_tensors="pt",
... )
>>> generated_ids = model.generate(**inputs)
>>> tokenizer.batch_decode(generated_ids, skip_special_tokens=True)
['Stan Lee created the character of Iron Man, a fictional superhero appearing in American comic']
```

## Citation