---
title: PISCES
canonical_url: "https://www.modelscope.cn/models/AI-ModelScope/PISCES"
md_url: "https://www.modelscope.cn/models/AI-ModelScope/PISCES.md"
repository: AI-ModelScope/PISCES
last_updated: 2024-12-31
license: cc-by-nc-sa-2.0
pipeline_tag: text2text-generation
tasks:
  - text2text-generation
model_type:
  - mbart
architectures:
  - MBartForConditionalGeneration
library_name:
  - pytorch
frameworks:
  - pytorch
downloads: 1065
stars: 0
---

# PISCES

> PISCES - AI-ModelScope 在 ModelScope 开源的模型。PISCES is a pre-trained many-to-many summarization model that learns language modeling, cross-lingual ability and summarization ability through a designed three-stage pre-training.

AI-ModelScope/PISCES 是 ModelScope 魔搭社区上的text2text-generation模型，采用 cc-by-nc-sa-2.0 许可。

- **Repository**: AI-ModelScope/PISCES
- **License**: cc-by-nc-sa-2.0
- **Tasks**: text2text-generation
- **Downloads**: 1065
- **Stars**: 0
- **Last updated**: 2024-12-31

Source: https://www.modelscope.cn/models/AI-ModelScope/PISCES

---

PISCES is a pre-trained many-to-many summarization model that learns language modeling, cross-lingual ability and summarization ability through a designed three-stage pre-training.

This model is introduced by *Towards Unifying Multi-Lingual and Cross-Lingual Summarization* (To appear in ACL 2023 main conference)


```python
tokenizer = MBart50Tokenizer.from_pretrained('Krystalan/PISCES')
model = MBartForConditionalGeneration.from_pretrained('Krystalan/PISCES')
```