# BART Paraphrasing
**URL:**
**Author:**  [@zanderbush](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fzanderbush)%5C
**Post date:** [July 15, 2020, 9:57pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbart-paraphrasing%2F312%2F1 "2020-07-15T21:57:55Z")
I’ve been using BART to summarize, and I have noticed some of the outputs resembling paraphrases.
Is there a way for me to build on this, and use the model for paraphrasing primarily?
```
from transformers import BartTokenizer, BartForConditionalGeneration, BartConfig
import torch
model = BartForConditionalGeneration.from_pretrained('facebook/bart-large-cnn')
tokenizer = BartTokenizer.from_pretrained('facebook/bart-large-cnn')
device = torch.device('cpu')
text = "At the core of the United States' mismanagement of the Coronavirus lies its distrust of science"
preprocess_text = text.strip().replace("\n","")
t5_prepared_Text = "summarize: "+preprocess_text
print ("original text preprocessed: \n", preprocess_text)
tokenized_text = tokenizer.encode(t5_prepared_Text, return_tensors="pt").to(device)
summary_ids = model.generate(tokenized_text,
num_beams=10,
no_repeat_ngram_size=1,
min_length=10,
num_return_sequences = 2,
max_length=20,
top_k = 100,
early_stopping=True)
output = tokenizer.decode(summary_ids[0], skip_special_tokens=True)
output1 = tokenizer.decode(summary_ids[1], skip_special_tokens=True)
```
`Summarized Text: The United States' mismanagement of the Coronavirus is rooted in its distrust of science.`
I’d like to note that when I do “num\_return\_sequences” the answers are the same. That makes sense, but is there a way for me to get separate answers? I don’t believe seed to be built-in with BART.
---
**Author:**  [@valhalla](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fvalhalla)%5C
**Post date:** [July 16, 2020, 5:32am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbart-paraphrasing%2F312%2F2 "2020-07-16T05:32:07Z")
hi @zanderbush, sure BART should also work for paraphrasing. Just fine-tune it on a paraphrasing dataset.
There’s a small mistake in the way you are using `.generate`. If you want to do sampling you’ll need to set `num_beams` to 0 and and `do_sample` to `True` . And set `do_sample` to false and `num_beams` to \>1 for beam search. This [post](https://huggingface.co/blog/how-to-generate) explains how to use `generate`
---
**Author:**  [@zanderbush](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fzanderbush)%5C
**Post date:** [July 16, 2020, 8:08pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbart-paraphrasing%2F312%2F3 "2020-07-16T20:08:12Z")
Thank you! I appreciate your help. I hope I am not probing too much, but I want to make sure I am doing this project in an efficient manner. Would T5 be better suited for paraphrasing?
---
**Author:**  [@valhalla](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fvalhalla)%5C
**Post date:** [July 17, 2020, 4:26am UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbart-paraphrasing%2F312%2F4 "2020-07-17T04:26:27Z")
I’ve only tried T5 for paraphrasing, but BART should also work, you’ll need to experiment and see what works best for your goal. Here’s a T5 paraphrasing [project](https://madewithml.com/projects/1094/paraphrase-any-question-with-t5-text-to-text-transformer/)
---
**Author:**  [@Guen](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2FGuen)%5C
**Post date:** [November 29, 2021, 7:58pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbart-paraphrasing%2F312%2F5 "2021-11-29T19:58:05Z")
Hello, @zanderbush, do you have an example of how you implemented the training and data pre-processing for this task? I am looking around, but I’m struggling to find good examples on the topic - thank you in advance!
---
**Author:**  [@jbmaxwell](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fjbmaxwell)%5C
**Post date:** [February 18, 2022, 6:06pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbart-paraphrasing%2F312%2F6 "2022-02-18T18:06:44Z")
Jumping on this thread a bit, I’m wondering whether a form of paraphrasing might be possible just by doing within-language “translation” and using sampling (top\_k, top\_p, temperature) to reduce the likelihood of simply reproducing exact quotations. Does that make sense, in lieu of an actual paraphrasing dataset? I’m asking because I have a use case where a dataset does not exist.
Thanks in advance.
---
**Author:**  [@jbmaxwell](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Fu%2Fjbmaxwell)%5C
**Post date:** [February 18, 2022, 6:09pm UTC](/static-proxy?url=https%3A%2F%2Fdiscuss.huggingface.co%2Ft%2Fbart-paraphrasing%2F312%2F7 "2022-02-18T18:09:13Z")
Oh, also @valhalla, would it be possible to re-post the T5 paraphrasing project? That link has died.