A Deep Relevance Matching Model for Ad-hoc Retrieval

By Jiafeng Guo, Yixing Fan, Qingyao Ai, W. Bruce Croft. 2016

944 citations recorded by the source corpus.

In recent years, deep neural networks have led to exciting breakthroughs in speech recognition, computer vision, and natural language processing (NLP) tasks. However, there have been few positive results of deep models on ad-hoc retrieval tasks. This is partially due to the fact that many important characteristics of the ad-hoc retrieval task have not been well addressed in deep models yet. Typically, the ad-hoc retrieval task is formalized as a matching problem between two pieces of text in existing work using deep models, and treated equivalent to many NLP tasks such as paraphrase identification, question answering and automatic conversation. However, we argue that the ad-hoc retrieval task is mainly about relevance matching while most NLP matching tasks concern semantic matching, and there are some fundamental differences between these two matching tasks. Successful relevance matching requires proper handling of the exact matching signals, query term importance, and diverse matching requirements. In this paper, we propose a novel deep relevance matching model (DRMM) for ad-hoc retrieval. Specifically, our model employs a joint deep architecture at the query term level for relevance matching. By using matching histogram mapping, a feed forward matching network, and a term gating network, we can effectively deal with the three relevance matching factors mentioned above. Experimental results on two representative benchmark collections show that our model can significantly outperform some well-known retrieval models as well as state-of-the-art deep matching models.

Subfields: Artificial Intelligence, Information Retrieval & Search

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 Information Retrieval & Search

Citations in the corpus

End-to-End Neural Ad-hoc Ranking with Kernel Pooling (2017, methodology): “For example, polynomial kernel can be used, but histograms [9] cannot as they are not dierentiable.”
Word-Entity Duet Representations for Document Ranking (2017, methodology): “en the histogram pooling technique is used to gather querydocument matching signals from T [9, 23]:”
Passage Re-ranking with BERT (2019, methodology): “The information retrieval (IR) community has also experienced a ﬂourishing development of neural ranking models, such as DRMM (Guo et al., 2016), KNRM (Xiong et al., 2017), Co-PACRR (Hui et al., 2018), and DUET (Mitra et al., 2017).”
Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrieval (2020, background): “eural information retrieval (Neu-IR) (Mitra et al., 2018), a common belief was that the interaction models, those that speciﬁcally handle term level matches, are more effective though more expensive (Guo et al., 2016; Xiong et al., 2017; Nogueira & Cho, 2019). Many techniques are developed to re”
ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT (2020, background): “Over the past few years, the Information Retrieval (IR) community has witnessed the introduction of a host of neural ranking models, including DRMM [8], KNRM [5, 36], and Duet [21, 23].”
Revealing the Importance of Semantic Retrieval for Machine Reading at Scale (2019, background): “Information Retrieval Success in deep neural networks inspires their application to information retrieval (IR) tasks (Huang et al., 2013; Guo et al., 2016; Mitra et al., 2017; Dehghani et al., 2017).”
Pretrained Transformers for Text Ranking: BERT and Beyond (2020, methodology): “nce signals from the similarity matrix. Conceptually, these represent different attempts to go beyond exact term matching using continuous vector representations of texts.27 Unigram models like DRMM [Guo et al., 2016] and KNRM [Xiong et al., 2017] aggregate the similarity between each query term and”
Sparse, Dense, and Attentional Representations for Text Retrieval (2020, neutral): “Prior and concurrent work has also used hybrid sparse-dense models (Guo et al., 2016a; Seo et al., 2019; Karpukhin et al., 2020; Ma et al., 2020; Gao et al., 2020).”
Long Range Arena: A Benchmark for Efficient Transformers (2020, background): “Learning the similarity score between two vectors is a common problem in machine learning and is useful for a wide array of applications (Guo et al., 2016).”
Context-Aware Sentence/Passage Term Importance Estimation For First Stage Retrieval (2019, methodology): “[13] used a similar word-embedding based query term weighting method in neural re-rankers.”
Multi-Stage Document Ranking with BERT (2019, methodology): “Well-known neural ranking models include DRMM (Guo et al., 2016), DUET (Mitra et al., 2017), KNRM (Xiong et al., 2017), and Co-PACRR (Hui et al., 2018); the literature is too vast for an exhaustive review here, and thus we refer readers to recent overviews (Onal et al., 2018; Mitra and Craswell,…”
Learning to Match using Local and Distributed Representations of Text for Web Search (2016, methodology): “While some negative results have been reported for title-based DSSM and CDSSM on the ad hoc document retrieval tasks [12, 30], we included document-based variants appropriately retrained on the same set of positive query and document pairs as our model.”
Simple Applications of BERT for Ad Hoc Document Retrieval (2019, methodology): “Researchers have understood for a few years now that relevance matching and semantic matching (for example, paraphrase detection, nat-ural language inference, etc.) are different tasks, despite shared common characteristics (Guo et al., 2016).”
Understanding the Behaviors of BERT in Ranking (2019, background): “These techniques lead to promising performances on various ranking benchmarks [1, 3, 5, 7].”
Neural Ranking Models with Weak Supervision (2017, methodology): “ut of network. For instance, DeepMatch [26] maps each text to a sequence of terms and trains a feed-forward network for computing the matching score. „e deep relevance matching modelforad-hocretrieval[14]isanotherexampleofanearlycombination model that feeds a neural network with the histogram-based ”
RocketQA: An Optimized Training Approach to Dense Passage Retrieval for Open-Domain Question Answering (2020, methodology): “Based on the retrieved passages from a first-stage traditional retriever, a number of (pre-BERT) neural models, such as DRMM (Guo et al., 2016), KNRM (Xiong et al.”
Context-Aware Document Term Weighting for Ad-Hoc Search (2020, background): “Being bag-of-words (BoW) representation, HDCT shares the advantages of classic bag-of-words document representations: efficient retrieval, support for fine-grained term signals [16], and higher interpretability compared to latent topic models and embeddings.”