# RAG for Reading Comprehension

**URL:** <https://discuss.huggingface.co/t/rag-for-reading-comprehension/5316>\
**Category:** Models\
**Created:** [April 5, 2021, 8:57pm UTC](https://discuss.huggingface.co/t/rag-for-reading-comprehension/5316 "2021-04-05T20:57:50Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![rmurthy](https://avatars.discourse-cdn.com/v4/letter/r/dc4da7/32.png) [@rmurthy](https://discuss.huggingface.co/u/rmurthy)\
**Post date:** [April 5, 2021, 8:57pm UTC](https://discuss.huggingface.co/t/rag-for-reading-comprehension/5316/1 "2021-04-05T20:57:50Z")

</div>

Hi,

I am currently working on a Reading Comprehension task. I was thinking of using RAG for answer _generation_ instead of using a span extraction model.

Find the starter code, question & passage below -

**query** = “Who is Adam’s sister?”  
**passage** = “Adam is Bob’s friend. Bob was born in 1906. Bob married Angela and they are now happily living together. Angela is Adam’s sister. Angela lives in Los Angeles. Bob has a dog and its name is Moxie. Adam likes Bob because Bob is a kind person. Adam has 2 kids.”

```
import torch
from transformers import RagConfig, RagRetriever, RagTokenForGeneration, RagTokenizer, RagSequenceForGeneration

tokenizer = RagTokenizer.from_pretrained("facebook/rag-token-nq")
nq_model = RagTokenForGeneration.from_pretrained("facebook/rag-token-nq")
nq_model.rag.config.n_docs = 1
_ = nq_model.to("cuda:0")
_ = nq_model.eval()

num_beams=1
min_length=10
max_length=16

device = "cuda:0"
input_ids = tokenizer(query, return_tensors="pt").input_ids.to(device)
passage_ids = tokenizer(passage, return_tensors="pt").input_ids.to(device)
passage_attention_mask = tokenizer(passage, return_tensors="pt").attention_mask.to(device)

generated_ids = nq_model.generate(
    input_ids=input_ids,
    context_input_ids=passage_ids,
    context_attention_mask=passage_attention_mask,
    doc_scores=torch.tensor([[100.0]]).to("cuda:0"),
    num_beams=1,
    num_return_sequences=1,
    min_length=min_length,
    max_length=max_length,
    length_penalty=1.0,
)
answer_texts = [
    tokenizer.generator.decode(gen_seq.tolist(), skip_special_tokens=True).strip() for gen_seq in generated_ids
]

print(answer_texts)

```

the model outputs **‘exit bar houses j j j j j’.**

Any thoughts on what might be wrong here?  
Thanks!

---

<div class="post-metadata">

**Author:** ![Jung](https://sea2.discourse-cdn.com/hellohellohello/user_avatar/discuss.huggingface.co/jung/32/21989_2.png) [@Jung](https://discuss.huggingface.co/u/Jung)\
**Post date:** [April 6, 2021, 2:48pm UTC](https://discuss.huggingface.co/t/rag-for-reading-comprehension/5316/2 "2021-04-06T14:48:08Z")

</div>

Maybe related: the latest work of google AI found that often RAG does not use context at all.

> **[Progress and Challenges in Long-Form Open-Domain Question Answering](https://ai.googleblog.com/2021/03/progress-and-challenges-in-long-form.html)**
>
> Posted by Aurko Roy, Research Scientist, Google Research Open-domain long-form question answering (LFQA) is a fundamental challenge in n...