mlsanityData to 7 Oct 2026

Optimizing Test-Time Query Representations for Dense Retrieval

arXiv 2205.12680 ↗First posted 25 May 2022cs.CL cs.IRcited by 6

Not on the map. Why?

Abstract (arXiv)

Recent developments of dense retrieval rely on quality representations of queries and contexts from pre-trained query and context encoders. In this paper, we introduce TOUR (Test-Time Optimization of Query Representations), which further optimizes instance-level query representations guided by signals from test-time retrieval results. We leverage a cross-encoder re-ranker to provide fine-grained pseudo labels over retrieval results and iteratively optimize query representations with gradient descent. Our theoretical analysis reveals that TOUR can be viewed as a generalization of the classical Rocchio algorithm for pseudo relevance feedback, and we present two variants that leverage pseudo-labels as hard binary or soft continuous labels. We first apply TOUR on phrase retrieval with our proposed phrase re-ranker, and also evaluate its effectiveness on passage retrieval with an off-the-shelf re-ranker. TOUR greatly improves end-to-end open-domain question answering accuracy, as well as passage retrieval performance. TOUR also consistently improves direct re-ranking by up to 2.0% while running 1.3-2.4x faster with an efficient implementation.

Key facts. Cited by 6 papers in the map (data through 7 Oct 2026). Builds on 34 of its 46 references that are in the map. First posted on arXiv 25 May 2022.

Builds on

34 of 46 references.

References · 34

  1. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
  2. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,115
  3. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,779
  4. 2017TriviaQA: A Large Scale Distantly Supervised Challenge Dataset for Reading Comprehensioncited by 2,895
  5. 2019The Curious Case of Neural Text Degenerationcited by 2,881
  6. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,126
  7. 2021BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Modelscited by 1,225
  8. 2020Leveraging Passage Retrieval with Generative Models for Open Domain Question Answeringcited by 1,203
  9. 2020ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERTcited by 1,160
  10. 2020Approximate Nearest Neighbor Negative Contrastive Learning for Dense Text Retrievalcited by 846
  11. 2017Reading Wikipedia to Answer Open-Domain Questionscited by 777
  12. 2019Latent Retrieval for Weakly Supervised Open Domain Question Answeringcited by 691
22 more
  1. 2019Passage Re-ranking with BERTcited by 629
  2. 2021Efficiently Teaching an Effective Dense Retriever with Balanced Topic Aware Samplingcited by 238
  3. 2020Distilling Knowledge from Reader to Retriever for Question Answeringcited by 226
  4. 2021PAQ: 65 Million Probably-Asked Questions and What You Can Do With Themcited by 198
  5. 2020Generation-Augmented Retrieval for Open-domain Question Answeringcited by 186
  6. 2021RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-rankingcited by 175
  7. 2021Simple Entity-Centric Questions Challenge Dense Retrieverscited by 140
  8. 2020Question and Answer Test-Train Overlap in Open-Domain Question Answering Datasetscited by 137
  9. 2021End-to-End Training of Multi-Document Reader and Retriever for Open-Domain Question Answeringcited by 133
  10. 2020Learning Dense Representations of Phrases at Scalecited by 91
  11. 2020Relevance-guided Supervision for OpenQA with ColBERTcited by 71
  12. 2019Real-Time Open-Domain Question Answering with Dense-Sparse Phrase Indexcited by 64
  13. 2021Learning to Retrieve Passages without Supervisioncited by 46
  14. 2021Improving Query Representations for Dense Retrieval with Pseudo Relevance Feedbackcited by 45
  15. 2021UnitedQA: A Hybrid Approach for Open Domain Question Answeringcited by 42
  16. 2021R2-D2: A Modular Baseline for Open-Domain Question Answeringcited by 37
  17. 2021Phrase Retrieval Learns Passage Retrieval, Toocited by 33
  18. 2021Pseudo-Relevance Feedback for Multiple Representation Dense Retrievalcited by 31
  19. 2021Pseudo Relevance Feedback with Deep Language Models and Dense Retrievers: Successes and Pitfallscited by 26
  20. 2019Dynamic Evaluation of Transformer Language Modelscited by 23
  21. 2018NPRF: A Neural Pseudo Relevance Feedback Framework for Ad-hoc Information Retrievalcited by 16
  22. 2021Boosted Dense Retrievercited by 10

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 2; 2025: 2; 2026: 2.

6 papers.

  1. 2025Guided Query Refinement: Multimodal Hybrid Retrieval with Test-Time Optimizationcited by 10
  2. 2025Test-time Corpus Feedback: From Retrieval to RAGcited by 3
  3. 2024Adaptive Retrieval and Scalable Indexing for k-NN Search with Cross-Encoderscited by 1
  4. 2026Document Optimization for Black-Box Retrieval via Reinforcement Learningcited by 1
  5. 2024SUKHSANDESH: An Avatar Therapeutic Question Answering Platform for Sexual Education in Rural Indiacited by 0
  6. 2026Task-Adaptive Embedding Refinement via Test-time LLM Guidancecited by 0