# DeepRank: A New Deep Architecture for Relevance Ranking in Information Retrieval

arXiv: https://arxiv.org/abs/1710.05649 · HTML: https://mlsanity.com/paper/1710.05649/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Oct 2017.

## Built on by (most cited first, 21 of 21)

- [Information Retrieval Meets Large Language Models: A Strategic Report from Chinese IR Community](https://mlsanity.com/paper/2307.09751/) (2023)
- [B-PROP: Bootstrapped Pre-training with Representative Words Prediction for Ad-hoc Retrieval](https://mlsanity.com/paper/2104.09791/) (2021)
- [Neural ranking models for document retrieval](https://mlsanity.com/paper/2102.11903/) (2021)
- [Overview of BioASQ 2020: The eighth BioASQ challenge on Large-Scale Biomedical Semantic Indexing and Question Answering](https://mlsanity.com/paper/2106.14618/) (2021)
- [Match-Prompt: Improving Multi-task Generalization Ability for Neural Text Matching via Prompt Learning](https://mlsanity.com/paper/2204.02725/) (2022)
- [RANK-NOSH: Efficient Predictor-Based Architecture Search via Non-Uniform Successive Halving](https://mlsanity.com/paper/2108.08019/) (2021)
- [RankDNN: Learning to Rank for Few-shot Learning](https://mlsanity.com/paper/2211.15320/) (2022)
- [Match-Ignition: Plugging PageRank into Transformer for Long-form Text Matching](https://mlsanity.com/paper/2101.06423/) (2021)
- [RankNEAT: Outperforming Stochastic Gradient Search in Preference Learning Tasks](https://mlsanity.com/paper/2204.06901/) (2022)
- [Revealing data leakage in protein interaction benchmarks](https://mlsanity.com/paper/2404.10457/) (2024)
- [AffRankNet+: Ranking Affect Using Privileged Information](https://mlsanity.com/paper/2108.05598/) (2021)
- [Query-Aware Graph Neural Networks for Enhanced Retrieval-Augmented Generation](https://mlsanity.com/paper/2508.05647/) (2025)
- [A Neural Model for Joint Document and Snippet Ranking in Question Answering for Large Document Collections](https://mlsanity.com/paper/2106.08908/) (2021)
- [Modeling Relevance Ranking under the Pre-training and Fine-tuning Paradigm](https://mlsanity.com/paper/2108.05652/) (2021)
- [YES SIR!Optimizing Semantic Space of Negatives with Self-Involvement Ranker](https://mlsanity.com/paper/2109.06436/) (2021)
- [Born for Auto-Tagging: Faster and better with new objective functions](https://mlsanity.com/paper/2206.07264/) (2022)
- [Effective and Efficient Query-aware Snippet Extraction for Web Search](https://mlsanity.com/paper/2210.08809/) (2022)
- [CSRM-LLM: Embracing Multilingual LLMs for Cold-Start Relevance Matching in Emerging E-commerce Markets](https://mlsanity.com/paper/2509.01566/) (2025)
- [Predicting Movie Hits Before They Happen with LLMs](https://mlsanity.com/paper/2505.02693/) (2025)
- [Detecting Mislabeled and Corrupted Data via Pointwise Mutual Information](https://mlsanity.com/paper/2508.07713/) (2025)
- [CausalSent: Interpretable Sentiment Classification with RieszNet](https://mlsanity.com/paper/2508.17576/) (2025)

## Abstract (arXiv)

This paper concerns a deep learning approach to relevance ranking in information retrieval (IR). Existing deep IR models such as DSSM and CDSSM directly apply neural networks to generate ranking scores, without explicit understandings of the relevance. According to the human judgement process, a relevance label is generated by the following three steps: 1) relevant locations are detected, 2) local relevances are determined, 3) local relevances are aggregated to output the relevance label. In this paper we propose a new deep learning architecture, namely DeepRank, to simulate the above human judgment process. Firstly, a detection strategy is designed to extract the relevant contexts. Then, a measure network is applied to determine the local relevances by utilizing a convolutional neural network (CNN) or two-dimensional gated recurrent units (2D-GRU). Finally, an aggregation network with sequential integration and term gating mechanism is used to produce a global relevance score. DeepRank well captures important IR characteristics, including exact/semantic matching signals, proximity heuristics, query term importance, and diverse relevance requirement. Experiments on both benchmark LETOR dataset and a large scale clickthrough data show that DeepRank can significantly outperform learning to ranking methods, and existing deep learning methods.