Bridging the Gap Between Indexing and Retrieval for Differentiable Search Index with Query Generation

By Shengyao Zhuang, Houxing Ren, Linjun Shou, Jian Pei, Ming Gong, Guido Zuccon, Daxin Jiang. 2022

89 citations recorded by the source corpus.

The Differentiable Search Index (DSI) is an emerging paradigm for information retrieval. Unlike traditional retrieval architectures where index and retrieval are two different and separate components, DSI uses a single transformer model to perform both indexing and retrieval. In this paper, we identify and tackle an important issue of current DSI models: the data distribution mismatch that occurs between the DSI indexing and retrieval processes. Specifically, we argue that, at indexing, current DSI methods learn to build connections between the text of long documents and the identifier of the documents, but then retrieval of document identifiers is based on queries that are commonly much shorter than the indexed documents. This problem is further exacerbated when using DSI for cross-lingual retrieval, where document text and query text are in different languages. To address this fundamental problem of current DSI models, we propose a simple yet effective indexing framework for DSI, called DSI-QG. When indexing, DSI-QG represents documents with a number of potentially relevant queries generated by a query generation model and re-ranked and filtered by a cross-encoder ranker. The presence of these queries at indexing allows the DSI models to connect a document identifier to a set of queries, hence mitigating data distribution mismatches present between the indexing and the retrieval phases. Empirical results on popular mono-lingual and cross-lingual passage retrieval datasets show that DSI-QG significantly outperforms the original DSI model.

Subfields: Computer Vision and Pattern Recognition, Information Systems, Retrieval-Augmented Generation, Natural Language Processing, Signal Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Ji-rong Wen
2. Karthik Narasimhan
3. Shunyu Yao
4. Wayne Xin Zhao
5. Ruiyang Ren
6. Howard Chen
7. John Yang
8. Jing Liu

In Information Systems

1. Ji-rong Wen
2. Wayne Xin Zhao
3. Karthik Narasimhan
4. Ruiyang Ren
5. Jing Liu
6. Yupeng Hou

In Retrieval-Augmented Generation

1. Ji-rong Wen
2. Ruiyang Ren
3. Wayne Xin Zhao
4. Jing Liu

In Natural Language Processing

1. Shunyu Yao
2. Karthik Narasimhan
3. John Yang
4. Howard Chen
5. Ji-rong Wen
6. Ruiyang Ren
7. Wayne Xin Zhao
8. Jing Liu

In Signal Processing

1. Ji-rong Wen
2. Ruiyang Ren
3. Wayne Xin Zhao
4. Jing Liu

Citations in the corpus

• WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agents (2022, background): “These include: 1) generation of good search queries [22, 59] and reformulation [37, 51], 2) strategic exploration for navigating through the website [55, 56, 29], 3) robust language understanding for textual state and action spaces [3, 7, 17, 44], and 4) long-term”
• Dense Text Retrieval Based on Pretrained Language Models: A Survey (2022, methodology): “Besides, since the amount of available relevance judgement data is usually limited, query generation is useful in these generative retrieval models for enhancing the learning from queries to docids [304], [305], where the original content of a document can be employed to generate potential queries.”