UnitedQA: A Hybrid Approach for Open Domain Question Answering

By Hao Cheng, Yelong Shen, Xiaodong Liu, Pengcheng He, Weizhu Chen, Jianfeng Gao. 2021

65 citations recorded by the source corpus.

To date, most of recent work under the retrieval-reader framework for open-domain QA focuses on either extractive or generative reader exclusively. In this paper, we study a hybrid approach for leveraging the strengths of both models. We apply novel techniques to enhance both extractive and generative readers built upon recent pretrained neural language models, and find that proper training methods can provide large improvement over previous state-of-the-art models. We demonstrate that a simple hybrid approach by combining answers from both readers can efficiently take advantages of extractive and generative answer inference strategies and outperforms single models as well as homogeneous ensembles. Our approach outperforms previous state-of-the-art models by 3.3 and 2.7 points in exact match on NaturalQuestions and TriviaQA respectively.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing

How researchers see this paper

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

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

Citations in the corpus

WebGPT: Browser-assisted question-answering with human feedback (2021, neutral): “Our results are shown in Table 9, along with those of the best existing model, UnitedQA [Cheng et al., 2021].”
You Only Need One Model for Open-domain Question Answering (2021, background): “UnitedQA (Cheng et al., 2021) and R2D2 (Fajcik et al., 2021) combine re-sults from an ensemble of extractive and generative readers, whereas PAQ (Lewis et al., 2021) directly retrieves answers with an FiD fallback.”
R2-D2: A Modular Baseline for Open-Domain Question Answering (2021, methodology): “Concurrently with our work, Cheng et al. (2021) proposed hard voting ensembling scheme to combine the reader predictions.”
Teaching language models to support answers with verified quotes (2022, background): “Generative language models (LMs) are increasingly useful for answering questions about the world, steadily improving (Cheng et al., 2021; Nakano et al., 2021; Roberts et al., 2020) on questionanswering benchmarks (Kwiatkowski et al.”
Generate rather than Retrieve: Large Language Models are Strong Context Generators (2022, background): “, 2022) or the reader (Izacard & Grave, 2021; Cheng et al., 2021; Yu et al., 2022), or training the system end-to-end (Lewis et al.”
The Power of Noise: Redefining Retrieval for RAG Systems (2024, unknown): “The idea had first originated in works such as [6] and [47], but the concept of RAG was popularized in [19], which introduced a model that combined a dense passage retriever with a sequence-to-sequence model,showcasingsubstantialimprovementsinknowledge-intensive tasks.”
LongRAG: Enhancing Retrieval-Augmented Generation with Long-context LLMs (2024, background): “…et al., 2020; Xiong et al., 2020a; Qu et al., 2020; Xiong et al., 2020b; Khalifa et al., 2023), enhancing the reader (Izacard and Grave, 2020b; Cheng et al., 2021; Yu et al., 2021; Borgeaud et al., 2022), fine-tuning the re-triever and reader jointly (Yu, 2022; Izacard et al., 2022; Singh et…”