Abstract:Humans are accustomed to autonomously associating prior knowledge with the text in a query when answering questions. However, machines lacking cognition and common sense, a query is merely a combination of some words. Although we can enrich the semantic information of the given query through language representation or query expansion (QE), the information contained in the query is still insufficient. In this paper, we propose an effective passage retrieval method named query context expansion-based retrieval (QCER) for open-domain question answering (OpenQA). QCER associates a query with domain information by adding contextual association information based on the pseudo-relevance feedback (PRF). QCER uses a dense reader to select top-n expansion terms for QE. We implement QCER by appending reader predictions, theoretically present in candidate passages, as contextual information to the initial query to form the new query. QCER with sparse representations (BM25) can improve retrieval efficiency and accelerate query convergence so that the reader can find the desired answer using fewer relevant passages, e.g., 10 passages, as soon as possible. Moreover, QCER can be easily combined with dense passage retrieval (DPR) to achieve even better performance, as sparse and dense representations are often complementary. Remarkably, we demonstrate that QCER achieves state-of-the-art performance in three tasks, passage retrieval, passage reading, and passage reranking, on the Natural Questions (NQ) and TriviaQA (Trivia) datasets under an extractive QA setup.

BERT-QE: Contextualized Query Expansion for Document Re-ranking

CEQE: Contextualized Embeddings for Query Expansion

Query Expansion Using Contextual Clue Sampling with Language Models

Expand, Rerank, and Retrieve: Query Reranking for Open-Domain Question Answering.

Modeling Queries with Contextual Snippets for Information Retrieval.

A context-dependent approach to query expansion in information retrieval

Query Context Expansion for Open-Domain Question Answering

An Empirical Study on Query Expansion and Documentexpans Ion in Information Retrieval.

A Context-Aware BERT Retrieval Framework Utilizing Abstractive Summarization.

Corpus-Steered Query Expansion with Large Language Models

Enhance Code Search Via Reformulating Queries with Evolving Contexts.

A Bayesian network approach to context sensitive query expansion.

Event-Centric Query Expansion in Web Search

Query Expansion Based on Context of Document and Search Result

Query Expansion Based on Semantics and Statistics in Chinese Question Answering System

Query Expansion with Local Conceptual Word Embeddings in Microblog Retrieval

Progressive Query Expansion for Retrieval Over Cost-constrained Data Sources

Query Expansion with Statistical Machine Translation

Exploration of Query Context for Information Retrieval

Improving Query Expansion Using WordNet

A Hybrid Text Generation-Based Query Expansion Method for Open-Domain Question Answering