ABSTRACT
Although considerable attention has been given to neural ranking architectures recently, far less attention has been paid to the term representations that are used as input to these models. In this work, we investigate how two pretrained contextualized language models (ELMo and BERT) can be utilized for ad-hoc document ranking. Through experiments on TREC benchmarks, we find that several ex-sting neural ranking architectures can benefit from the additional context provided by contextualized language models. Furthermore, we propose a joint approach that incorporates BERT's classification vector into existing neural models and show that it outperforms state-of-the-art ad-hoc ranking baselines. We call this joint approach CEDR (Contextualized Embeddings for Document Ranking). We also address practical challenges in using these models for ranking, including the maximum input length imposed by BERT and runtime performance impacts of contextualized language models.
- Zhuyun Dai, Chenyan Xiong, James P. Callan, and Zhiyuan Liu. 2018. Convolutional Neural Networks for Soft-Matching N-Grams in Ad-hoc Search. In WSDM . Google ScholarDigital Library
- Mostafa Dehghani, Hamed Zamani, Aliaksei Severyn, Jaap Kamps, and W. Bruce Croft. 2017. Neural Ranking Models with Weak Supervision. In SIGIR . Google ScholarDigital Library
- Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. 2009. Imagenet: A large-scale hierarchical image database. In Computer Vision and Pattern Recognition .Google Scholar
- Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. arXiv preprint arXiv:1810.04805 (2018).Google Scholar
- Jiafeng Guo, Yixing Fan, Qingyao Ai, and W. Bruce Croft. 2016. A Deep Relevance Matching Model for Ad-hoc Retrieval. In CIKM . Google ScholarDigital Library
- Kai Hui, Andrew Yates, Klaus Berberich, and Gerard de Melo. 2018. Co-PACRR: A Context-Aware Neural IR Model for Ad-hoc Retrieval. In WSDM . Google ScholarDigital Library
- Samuel Huston and W Bruce Croft. 2014. Parameters learned in the comparison of retrieval models using term dependencies. Technical Report (2014).Google Scholar
- Diederik P. Kingma and Jimmy Ba. 2015. Adam: A Method for Stochastic Optimization. In ICLR .Google Scholar
- Kui-Lam Kwok, Laszlo Grunfeld, H. L. Sun, and Peter Deng. 2004. TREC 2004 Robust Track Experiments Using PIRCS. In TREC .Google Scholar
- Nut Limsopatham, Richard McCreadie, M-Dyaa Albakour, Craig MacDonald, Rodrygo L. T. Santos, and Iadh Ounis. 2012. University of Glasgow at TREC 2012: Experiments with Terrier. In TREC .Google Scholar
- Xitong Liu, Peilin Yang, and Hui Fang. 2014. Entity Came to Rescue - Leveraging Entities to Minimize Risks in Web Search. In TREC .Google Scholar
- Ryan McDonald, Yichun Ding, and Ion Androutsopoulos. 2018. Deep Relevance Ranking using Enhanced Document-Query Interactions. In EMNLP .Google Scholar
- Donald Metzler and W. Bruce Croft. 2005. A Markov random field model for term dependencies. In SIGIR . Google ScholarDigital Library
- Rodrigo Nogueira and Kyunghyun Cho. 2019. Passage Re-ranking with BERT. CoRR , Vol. abs/1901.04085 (2019).Google Scholar
- Jeffrey Pennington, Richard Socher, and Christopher D. Manning. 2014. Glove: Global Vectors for Word Representation. In EMNLP .Google Scholar
- Matthew E. Peters, Mark Neumann, Mohit Iyyer, Matt Gardner, Christopher Clark, Kenton Lee, and Luke Zettlemoyer. 2018. Deep contextualized word representations. In Proc. of NAACL .Google ScholarCross Ref
- Fiana Raiber and Oren Kurland. 2013. The Technion at TREC 2013 Web Track: Cluster-based Document Retrieval. In TREC .Google Scholar
- Corby Rosset, Damien Jose, Gargi Ghosh, Bhaskar Mitra, and Saurabh Tiwary. 2018. Optimizing Query Evaluations Using Reinforcement Learning for Web Search. In SIGIR . Google ScholarDigital Library
- Ashish Vaswani, Noam Shazeer, Niki Parmar, Jakob Uszkoreit, Llion Jones, Aidan N. Gomez, Lukasz Kaiser, and Illia Polosukhin. 2017. Attention Is All You Need. In NIPS . Google ScholarDigital Library
- Chenyan Xiong, Zhuyun Dai, James P. Callan, Zhiyuan Liu, and Russell Power. 2017. End-to-End Neural Ad-hoc Ranking with Kernel Pooling. In SIGIR . Google ScholarDigital Library
- Peilin Yang, Hui Fang, and Jimmy Lin. 2017. Anserini: Enabling the Use of Lucene for Information Retrieval Research. In SIGIR . Google ScholarDigital Library
- Wei Yang, Yuqing Xie, Aileen Lin, Xingyu Li, Luchen Tan, Kun Xiong, Ming Li, and Jimmy Lin. 2019. End-to-End Open-Domain Question Answering with BERTserini. CoRR , Vol. abs/1901.04085 (2019).Google Scholar
- Jason Yosinski, Jeff Clune, Yoshua Bengio, and Hod Lipson. 2014. How transferable are features in deep neural networks?. In NIPS . Google ScholarDigital Library
- Hamed Zamani, Mostafa Dehghani, Fernando Diaz, Hang Li, and Nick Craswell. 2018. SIGIR 2018 Workshop on Learning from Limited or Noisy Data for Information Retrieval. In SIGIR . Google ScholarDigital Library
Index Terms
- CEDR: Contextualized Embeddings for Document Ranking
Recommendations
Fast Item Ranking under Neural Network based Measures
WSDM '20: Proceedings of the 13th International Conference on Web Search and Data MiningRecently, plenty of neural network based recommendation models have demonstrated their strength in modeling complicated relationships between heterogeneous objects (i.e., users and items). However, the applications of these fine trained recommendation ...
Order-Disorder: Imitation Adversarial Attacks for Black-box Neural Ranking Models
CCS '22: Proceedings of the 2022 ACM SIGSAC Conference on Computer and Communications SecurityNeural text ranking models have witnessed significant advancement and are increasingly being deployed in practice. Unfortunately, they also inherit adversarial vulnerabilities of general neural models, which have been detected but remain underexplored ...
End-to-End Contextualized Document Indexing and Retrieval with Neural Networks
SIGIR '20: Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information RetrievalRelevance in ad-hoc retrieval is a fundamental problem of text understanding. Developing neural network methods for this foundational task of Information Retrieval (IR) has the potential to impact many search domains. Recently, a new generation of ...
Comments