Difference between revisions of "Dense Passage Retrieval for Open-Domain Question Answering"

From statwiki
Jump to: navigation, search
Line 3: Line 3:
  
 
= 1. Introduction =
 
= 1. Introduction =
Open domain question answering is a task that finds question answers from a large collection of documents. Nowadays open domain QA systems usually use a two-stage framework: (1) a ''retriever'' that selects a subset of documents, and (2) a ''reader'' that fully reads the document subset and selects the answer spans. Stage one (1) is usually done through bag-of-words models, which count overlapping words and their frequencies in documents. A common bag-of-words method that has been used for years is BM25, which ranks all documents based on the query terms appearing in each document. Stage one produces a small subset of documents where the answer might appear, and then in stage two, a reader would read the subset and locate the answer spans. Stage two is usually done through neural models, like Bert. While stage two benefits a lot from the recent advancement of language models, stage one still relies on traditional term-based models. This paper tries to improve stage one by using dense retrieval methods, and demonstrates that dense retrieval methods can not only outperform BM25, but also improve the end-to-end QA accuracies.  
+
Open domain question answering is a task that finds question answers from a large collection of documents. Nowadays open domain QA systems usually use a two-stage framework: (1) a ''retriever'' that selects a subset of documents, and (2) a ''reader'' that fully reads the document subset and selects the answer spans. Stage one (1) is usually done through bag-of-words models, which count overlapping words and their frequencies in documents. Each document is represented by a high-dimensional, sparse vector. A common bag-of-words method that has been used for years is BM25, which ranks all documents based on the query terms appearing in each document. Stage one produces a small subset of documents where the answer might appear, and then in stage two, a reader would read the subset and locate the answer spans. Stage two is usually done through neural models, like Bert. While stage two benefits a lot from the recent advancement of language models, stage one still relies on traditional term-based models. This paper tries to improve stage one by using dense retrieval methods that generate dense, latent semantic document embedding, and demonstrates that dense retrieval methods can not only outperform BM25, but also improve the end-to-end QA accuracies.  
  
 
= 2. Background =
 
= 2. Background =
 
+
The following example clearly shows what problems open domain QA systems tackle. Given a question: "What is Uranus?", a system should find the answer spans from a large corpus. The corpus size can be billions of documents. In stage one, a retriever would select a small set of potentially relevant documents, which then would be fed to a neural reader in stage two for the answer spans extraction. Only a filtered subset of documents is processed by a neural reader since neural reading comprehension is expensive. It's impractical to process billions of documents using a neural reader.
  
 
= 3. Dense Passage Retriever =
 
= 3. Dense Passage Retriever =

Revision as of 18:02, 14 November 2020

Presented by

Nicole Yan

1. Introduction

Open domain question answering is a task that finds question answers from a large collection of documents. Nowadays open domain QA systems usually use a two-stage framework: (1) a retriever that selects a subset of documents, and (2) a reader that fully reads the document subset and selects the answer spans. Stage one (1) is usually done through bag-of-words models, which count overlapping words and their frequencies in documents. Each document is represented by a high-dimensional, sparse vector. A common bag-of-words method that has been used for years is BM25, which ranks all documents based on the query terms appearing in each document. Stage one produces a small subset of documents where the answer might appear, and then in stage two, a reader would read the subset and locate the answer spans. Stage two is usually done through neural models, like Bert. While stage two benefits a lot from the recent advancement of language models, stage one still relies on traditional term-based models. This paper tries to improve stage one by using dense retrieval methods that generate dense, latent semantic document embedding, and demonstrates that dense retrieval methods can not only outperform BM25, but also improve the end-to-end QA accuracies.

2. Background

The following example clearly shows what problems open domain QA systems tackle. Given a question: "What is Uranus?", a system should find the answer spans from a large corpus. The corpus size can be billions of documents. In stage one, a retriever would select a small set of potentially relevant documents, which then would be fed to a neural reader in stage two for the answer spans extraction. Only a filtered subset of documents is processed by a neural reader since neural reading comprehension is expensive. It's impractical to process billions of documents using a neural reader.

3. Dense Passage Retriever

3.1 Model Architecture Overview

3.2 Training

4. Experimental Setup

5. Retrieval Performance Evaluation

5.1 Main Results

5.2 Ablation Study on Model Training

5.3 Qualitative Analysis

5.4 Run-time Efficiency

6. Experiments: Question Answering

7. Related Work

8. Conclusion

Critiques

References

[1] Vladimir Karpukhin, Barlas Oğuz, Sewon Min, Patrick Lewis, Ledell Wu, Sergey Edunov, Danqi Chen, Wen-tau Yih. Dense Passage Retrieval for Open-Domain Question Answering. EMNLP 2020.