Overview of the TREC 2019 deep learning track

Abstract

The Deep Learning Track is a new track for TREC 2019, with the goal of studying ad hoc ranking in a large data regime. It is the first track with large human-labeled training sets, introducing two sets corresponding to two tasks, each with rigorous TREC-style blind evaluation and reusable test sets. The document retrieval task has a corpus of 3.2 million documents with 367 thousand training queries, for which we generate a reusable test set of 43 queries. The passage retrieval task has a corpus of 8.8 million passages with 503 thousand training queries, for which we generate a reusable test set of 43 queries. This year 15 groups submitted a total of 75 runs, using various combinations of deep learning, transfer learning and traditional IR ranking methods. Deep learning runs significantly outperformed traditional IR runs. Possible explanations for this result are that we introduced large training data and we included deep models trained on such data in our judging pools, whereas some past studies did not have such training data or pooling.

References in corpus (3)

• XLNet: Generalized Autoregressive Pretraining for Language Understanding
• An Updated Duet Model for Passage Re-ranking
• Incorporating Query Term Independence Assumption for Efficient Retrieval and Ranking using Deep Neural Networks

Cited by in corpus (19)

• Semantic Models for the First-stage Retrieval: A Comprehensive Review
• Pretrained Transformers for Text Ranking: BERT and Beyond
• Beyond 512 Tokens: Siamese Multi-depth Transformer-based Hierarchical Encoder for Long-Form Document Matching
• The Information Retrieval Experiment Platform
• Fast Passage Re-ranking with Contextualized Exact Term Matching and Efficient Passage Expansion
• ABNIRML: Analyzing the Behavior of Neural IR Models
• Lexically-Accelerated Dense Retrieval
• mMARCO: A Multilingual Version of the MS MARCO Passage Ranking Dataset
• The Power of Selecting Key Blocks with Local Pre-ranking for Long Document Information Retrieval
• Mr. TyDi: A Multi-lingual Benchmark for Dense Retrieval
• LexBoost: Improving Lexical Document Retrieval with Nearest Neighbors
• Neural Passage Quality Estimation for Static Pruning
• Learning Passage Impacts for Inverted Indexes
• SPLADE: Sparse Lexical and Expansion Model for First Stage Ranking
• Duet at TREC 2019 Deep Learning Track
• Pre-training for Ad-hoc Retrieval: Hyperlink is Also You Need
• Conformer-Kernel with Query Term Independence at TREC 2020 Deep Learning Track
• TREC Deep Learning Track: Reusable Test Collections in the Large Data Regime
• MS MARCO: Benchmarking Ranking Models in the Large-Data Regime