2020

Leveraging Semantic and Lexical Matching to Improve the Recall of Document Retrieval Systems: A Hybrid Approach

Kuzi, Saar, Zhang, Mingyang, Li, Cheng et al.

Understand

Search engines often follow a two-phase paradigm where in the first stage (the retrieval stage) an initial set of documents is retrieved and in the second stage (the re-ranking stage) the documents are re-ranked to obtain the final result list.

While deep neural networks were shown to improve the performance of the re-ranking stage in previous works, there is little literature about using deep neural networks to improve the retrieval stage.
In this paper, we study the merits of combining deep neural network models and lexical models for the retrieval stage.
A hybrid approach, which leverages both semantic (deep neural network-based) and lexical (keyword matching-based) retrieval models, is proposed.

Built on

Transformer-xl: Attentive language models beyond a fixed-length context
Original
Zihang Dai, Zhilin Yang, Yiming Yang, William W Cohen, Jaime Carbonell, Quoc V Le, and Ruslan Salakhutdinov. 2019 · 1901
Earlier work this paper cites.
Passage Re-ranking with BERT
Original
Rodrigo Nogueira and Kyunghyun Cho. 2019 · 1901
Earlier work this paper cites.
Simple applications of BERT for ad hoc document retrieval
Original
Wei Yang, Haotian Zhang, and Jimmy Lin. 2019 · 1903
Earlier work this paper cites.
Context-aware sentence/passage term importance estimation for first stage retrieval
Original
Zhuyun Dai and Jamie Callan. 2019 · 1910
Earlier work this paper cites.
Multi-stage document ranking with BERT
Original
Rodrigo Nogueira, Wei Yang, Kyunghyun Cho, and Jimmy Lin. 2019b · 1910
Earlier work this paper cites.
Signature verification using a "siamese" time delay neural network. In Advances in Neural Information Processing Systems . 737–744
Jane Bromley, Isabelle Guyon, Yann LeCun, Eduard Säckinger, and Roopak Shah. 1994 · 1994
Earlier work this paper cites.
Some simple effective approximations to the 2-poisson model for probabilistic weighted retrieval. In Proc. of the 17th Annual International ACM SIGIR Conference on Research and Development in Information Retrieval . 232–241
Stephen E Robertson and Steve Walker. 1994 · 1994
Earlier work this paper cites.
Learned vector-space models for document retrieval
William R Caid, Susan T Dumais, and Stephen I Gallant. 1995 · 1995
Earlier work this paper cites.
UMass at TREC 2004: Novelty and HARD. In Proc. of the 13th Text Retrieval Conference . 13
Nasreen Abdul-Jaleel, James Allan, W Bruce Croft, Fernando Diaz, Leah Larkey, Xiaoyan Li, Mark D Smucker, and Courtney Wade. 2004 · 2004
Earlier work this paper cites.
Techniques for efficient query expansion. In International Symposium on String Processing and Information Retrieval . Springer, 30–42
Bodo Billerbeck and Justin Zobel. 2004 · 2004
Earlier work this paper cites.
Document expansion versus query expansion for ad-hoc retrieval. In Proceedings of the 10th Australasian Document Computing Symposium . Citeseer, 34–41
Bodo Billerbeck and Justin Zobel. 2005 · 2005
Earlier work this paper cites.
Efficient and self-tuning incremental query expansion for top-k query processing. In Proceedings of the 28th annual international ACM SIGIR conference on Research and development in information retrieval . 242–249
Martin Theobald, Ralf Schenkel, and Gerhard Weikum. 2005 · 2005
Earlier work this paper cites.
Real-time query expansion in relevance models
Victor Lavrenko and James Allan. 2006 · 2006
Earlier work this paper cites.
Language model information retrieval with document expansion. In Proceedings of the Human Language Technology Conference of the NAACL, Main Conference . 407–414
Tao Tao, Xuanhui Wang, Qiaozhu Mei, and ChengXiang Zhai. 2006 · 2006
Earlier work this paper cites.

Similar

Visualizing data using t-SNE
Laurens van der Maaten and Geoffrey Hinton. 2008 · 2008
Cited alongside, same era.
Latent semantic indexing (LSI) fails for TREC collections
Avinash Atreya and Charles Elkan. 2011 · 2011
Cited alongside, same era.
A survey of automatic query expansion in information retrieval
Claudio Carpineto and Giovanni Romano. 2012 · 2012
Cited alongside, same era.
Learning deep structured semantic models for web search using clickthrough data. In Proc. of the 22nd ACM International Conference on Information and Knowledge Management . 2333–2338
Po-Sen Huang, Xiaodong He, Jianfeng Gao, Li Deng, Alex Acero, and Larry Heck. 2013 · 2013
Cited alongside, same era.
Two-stage hashing for fast document retrieval. In Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers) . 495–500
Hao Li, Wei Liu, and Heng Ji. 2014 · 2014
Cited alongside, same era.
Scalable nearest neighbor algorithms for high dimensional data
Marius Muja and David G. Lowe. 2014 · 2014
Cited alongside, same era.
A latent semantic model with convolutional-pooling structure for information retrieval. In Proc. of the 23rd ACM International Conference on Information and Knowledge Management . 101–110
Yelong Shen, Xiaodong He, Jianfeng Gao, Li Deng, and Grégoire Mesnil. 2014 · 2014
Cited alongside, same era.
Off the beaten path: Let’s replace term-based retrieval with k-nn search. In Proceedings of the 25th ACM international on conference on information and knowledge management . 1099–1108
Leonid Boytsov, David Novak, Yury Malkov, and Eric Nyberg. 2016 · 2016
Cited alongside, same era.
Neural ranking models with weak supervision. In Proc. of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval . 65–74
Mostafa Dehghani, Hamed Zamani, Aliaksei Severyn, Jaap Kamps, and W Bruce Croft. 2017 · 2017
Cited alongside, same era.
Learning to match using local and distributed representations of text for web search. In Proceedings of the 26th International Conference on World Wide Web . 1291–1299
Bhaskar Mitra, Fernando Diaz, and Nick Craswell. 2017 · 2017
Cited alongside, same era.
End-to-end neural ad-hoc ranking with kernel pooling. In Proc. of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval . 55–64
Chenyan Xiong, Zhuyun Dai, Jamie Callan, Zhiyuan Liu, and Russell Power. 2017 · 2017
Cited alongside, same era.
Query expansion using local and global document analysis
Jinxi Xu and W Bruce Croft. 2017 · 2017
Cited alongside, same era.

Then

Multi-level abstraction convolutional model with weak supervision for information retrieval. In Proc. of the 41st International ACM SIGIR Conference on Research and Development in Information Retrieval . 985–988
Yifan Nie, Alessandro Sordoni, and Jian-Yun Nie. 2018 · 2018
Later among the works it cites.
Neural information retrieval: At the end of the early years
Kezban Dilek Onal, Ye Zhang, Ismail Sengor Altingovde, Md Mustafizur Rahman, Pinar Karagoz, Alex Braylan, Brandon Dang, Heng-Lu Chang, Henna Kim, Quinten McNamara, et al · 2018
Later among the works it cites.
From neural re-ranking to neural ranking: Learning a sparse representation for inverted indexing. In Proc. of the 27th ACM International Conference on Information and Knowledge Management . 497–506
Hamed Zamani, Mostafa Dehghani, W Bruce Croft, Erik Learned-Miller, and Jaap Kamps. 2018 · 2018
Later among the works it cites.
Asking clarifying questions in open-domain information-seeking conversations. In Proc. of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval . 475–484
Mohammad Aliannejadi, Hamed Zamani, Fabio Crestani, and W Bruce Croft. 2019 · 2019
Later among the works it cites.
Learning more from less: Towards strengthening weak supervision for ad-hoc retrieval. In Proc. of the 42nd International ACM SIGIR Conference on Research and Development in Information Retrieval . 857–860
Dany Haddad and Joydeep Ghosh. 2019 · 2019
Later among the works it cites.
Latent retrieval for weakly supervised open domain question answering. In Proc. of the 57th Annual Meeting of the Association for Computational Linguistics . 6086–6096
Kenton Lee, Ming-Wei Chang, and Kristina Toutanova. 2019 · 2019
Later among the works it cites.
Incorporating query term independence assumption for efficient retrieval and ranking using deep neural networks
Original
Bhaskar Mitra, Corby Rosset, David Hawking, Nick Craswell, Fernando Diaz, and Emine Yilmaz. 2019 · 2019
Later among the works it cites.
Semantic product search. In Proc. of the 25th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining . 2876–2885
Priyanka Nigam, Yiwei Song, Vijai Mohan, Vihan Lakshman, Weitian Ding, Ankit Shingavi, Choon Hui Teo, Hao Gu, and Bing Yin. 2019 · 2019
Later among the works it cites.
From doc2query to docTTTTTquery
Rodrigo Nogueira, Jimmy Lin, and AI Epistemic. 2019a · 2019
Later among the works it cites.
Document Expansion by Query Prediction
Original
Rodrigo Nogueira, Wei Yang, Jimmy Lin, and Kyunghyun Cho. 2019c · 2019
Later among the works it cites.
Cross-domain modeling of sentence-level evidence for document retrieval. In Proc. of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing . 3481–3487
Zeynep Akkalyoncu Yilmaz, Wei Yang, Haotian Zhang, and Jimmy Lin. 2019 · 2019
Later among the works it cites.
Overview of the trec 2019 deep learning track
Original
Nick Craswell, Bhaskar Mitra, Emine Yilmaz, Daniel Campos, and Ellen M Voorhees. 2020 · 2020
Closest in time.
Accelerating Large-Scale Inference with Anisotropic Vector Quantization. In Proc. of the 37th International Conference on Machine Learning
Ruiqi Guo, Philip Sun, Erik Lindgren, Quan Geng, David Simcha, Felix Chern, and Sanjiv Kumar. 2020 · 2020
Closest in time.

Beyond the bibliography

alphaXiv searches the wider corpus for related work and actual follow-ups.

Open on alphaXiv

alphaXiv is searching for related work…

Reading the bibliography…