Recasting Web-Scale Query Suggestion as dense retrieval: Efficient, Up-to-Date, and Context-Aware Suggestions
sigir26-170 · Full research paper · Sosuke Nishikawa, Naoki Yoshinaga, Nobuhiro Kaji
Conversational Search and Retrieval · 50% Query Reformulation and Expansion · 25% Dense Bi-Encoder Training and Repres · 12% Zero-Shot and Domain Generalization · 12%
Abstract
Query suggestion (QS) in web search must provide efficient and up-to-date suggestions that are relevant to the session context. Recent studies treat QS as generation, which captures session context well but suffers from high latency and costly retraining to maintain information freshness. In this study, we recast QS as dense retrieval: given a search session, the next query is retrieved from a large index of historical queries using efficient approximate nearest neighbor search. We propose Context-aware Asymmetric Dual Encoder for QS (CADE-QS), which uses an asymmetric dual encoder to model query -to- suggestion dependency and incorporates session history into the query encoder via context-aware contrastive learning. We evaluate our method on two real-world search-log datasets: a recent Japanese web search log and the public AOL log. CADE-QS rivals strong generative baselines in quality while reducing end-to-end latency to around 30 ms on CPU, representing an order-of-magnitude improvement. Detailed analyses confirm CADE-QS's robust context awareness, unidirectional modeling, and practicality for reflecting evolving information via index refreshes, as well as the effectiveness of an adaptive hybrid strategy for low-coverage scenarios. Our code is publicly available at https://github.com/lycorp-jp/cadeqs.
Assigned reviewers
| # | Reviewer | Match | Why (closest evidence publications in the taxonomy) | |
|---|---|---|---|---|
| 1 | Negar Arabzadeh | 59 | 59 Enhanced Retrieval Effectiveness through Selective Query Generation (2024) Query Reformulation and Expansion 59 ReFormeR: Learning and Applying Explicit Query Reformulation Patterns (2026) Query Reformulation and Expansion 59 Query Performance Prediction: From Ad-hoc to Conversational Search (2023) Query Performance Prediction |
|
| 2 | Sheng-Chieh Lin | 58 | 86 Contextualized Query Embeddings for Conversational Search (2021) Conversational Search and Retrieval 84 Multi-Stage Conversational Passage Retrieval: An Approach to Fusing Term Importance Estimation and Neural Query Rewriting (2021) Conversational Search and Retrieval 72 Chatty Goose: A Python Framework for Conversational Search (2021) Conversational Search and Retrieval |
|
| 3 | Corby Rosset | 50 | 72 Contextual Re-Ranking with Behavior Aware Transformers (2020) Conversational Search and Retrieval 71 Zero-shot Clarifying Question Generation for Conversational Search (2023) Conversational and Session Search 46 Generic Intent Representation in Web Search (2019) Text Embedding and Representation Learning |
Area chair
| Area chair | Match | |
|---|---|---|
| Kam-Fai Wong | 59 |
Area-chair candidates (7)
| Chenyan Xiong | 61 | 15 papers | |
| Kam-Fai Wong assigned | 59 | 6 papers | |
| Hongjin Qian | 58 | 12 papers | |
| Fernando Diaz 0001 | 49 | 12 papers | |
| Jun Xu 0001 | 49 | 2 papers | |
| Xiaoxue Zang | 47 | 4 papers | |
| Xiaoxi Li 0005 | 43 | 15 papers |
Candidates from the panel ranked by match score
| # | Reviewer | Match | Load | Why | |
|---|---|---|---|---|---|
| 1 | Xiao Wang 0043 | 62 | 3 | 67 Improving zero-shot retrieval using dense external expansion (2022) Query Reformulation and Expansion 61 Effective Contrastive Weighting for Dense Query Expansion (2023) Query Reformulation and Expansion |
|
| 3 | Shengyao Zhuang | 60 | 3 | 67 To Interpolate or not to Interpolate: PRF, Dense and Sparse Retrievers (2022) Query Reformulation and Expansion 61 Improving Query Representations for Dense Retrieval with Pseudo Relevance Feedback: A Reproducibility Study (2022) Query Reformulation and Expansion |
|
| 5 | Negar Arabzadeh assigned | 59 | 3 | 59 Enhanced Retrieval Effectiveness through Selective Query Generation (2024) Query Reformulation and Expansion 59 ReFormeR: Learning and Applying Explicit Query Reformulation Patterns (2026) Query Reformulation and Expansion |
|
| 6 | Sheng-Chieh Lin assigned | 58 | 3 | 86 Contextualized Query Embeddings for Conversational Search (2021) Conversational Search and Retrieval 84 Multi-Stage Conversational Passage Retrieval: An Approach to Fusing Term Importance Estimation and Neural Query Rewriting (2021) Conversational Search and Retrieval |
|
| 8 | Xueguang Ma | 55 | 3 | 65 To Interpolate or not to Interpolate: PRF, Dense and Sparse Retrievers (2022) Query Reformulation and Expansion 59 Improving Query Representations for Dense Retrieval with Pseudo Relevance Feedback: A Reproducibility Study (2022) Query Reformulation and Expansion |
|
| 9 | Idan Szpektor | 53 | 3 | 69 On the Robustness of Dialogue History Representation in Conversational Question Answering: A Comprehensive Study and a New Prompt-based Method (2023) Conversational Search and Retrieval 65 When relevance is not enough: promoting diversity and freshnessin personalized question recommendation (2013) Conversational and Interactive Recommendation |
|
| 10 | Jingtao Zhan | 50 | 3 | 65 Capability-aware Prompt Reformulation Learning for Text-to-Image Generation (2024) Conversational Search and Retrieval 51 Combining Multiple Supervision for Robust Zero-Shot Dense Retrieval (2024) Sparse and Dense Retrieval Encoders |
|
| 11 | Corby Rosset assigned | 50 | 3 | 72 Contextual Re-Ranking with Behavior Aware Transformers (2020) Conversational Search and Retrieval 71 Zero-shot Clarifying Question Generation for Conversational Search (2023) Conversational and Session Search |
|
| 13 | Xinyu Ma 0001 | 49 | 3 | 48 Pre-training with Aspect-Content Text Mutual Prediction for Multi-Aspect Dense Retrieval (2023) Sparse and Dense Retrieval Encoders 48 Pre-train a Discriminative Text Encoder for Dense Retrieval via Contrastive Span Prediction (2022) Sparse and Dense Retrieval Encoders |
|
| 14 | Daiting Shi | 49 | 3 | 59 Population-Guided Intent-Aware Query Rewriting for Web Search (2026) Query Reformulation and Expansion 56 Accurate and Efficient Personalized Query Rewriting in Baidu Search (2026) Query Reformulation and Expansion |
|
| 15 | Ruqing Zhang 0001 | 49 | 3 | 48 B-PROP: Bootstrapped Pre-training with Representative Words Prediction for Ad-hoc Retrieval (2021) Sparse and Dense Retrieval Encoders 48 PROP: Pre-training with Representative Words Prediction for Ad-hoc Retrieval (2021) Sparse and Dense Retrieval Encoders |
|
| 16 | Gustavo Penha | 49 | 3 | 73 Pairwise Review-Based Explanations for Voice Product Search (2022) Conversational Search and Retrieval 69 On the Calibration and Uncertainty of Neural Learning to Rank Models for Conversational Search (2021) Conversational Search and Retrieval |
|
| 18 | Jialong Wu 0007 | 49 | 3 | 86 AdaRewriter: Unleashing the Power of Prompting-based Conversational Query Reformulation via Test-Time Adaptation (2025) Query Reformulation and Expansion 86 AdaCQR: Enhancing Query Reformulation for Conversational Search via Sparse and Dense Retrieval Alignment (2025) Query Reformulation and Expansion |
|
| 19 | Jurek Leonhardt | 49 | 3 | 53 Understanding the User: An Intent-Based Ranking Dataset (2024) Query Reformulation and Expansion 50 Efficient Neural Ranking Using Forward Indexes and Lightweight Encoders (2024) Learned Sparse Retrieval Efficiency |
|
| 20 | Falk Scholer | 48 | 3 | 59 Query association for effective retrieval (2002) Query Reformulation and Expansion 57 Ranking Documents by Answer-Passage Quality (2018) Retrieval-Augmented and Open-Domain QA |
|
| 22 | Wonduk Seo | 47 | 3 | 86 A New Query Expansion Approach for Enhancing Information Retrieval via Agent-Mediated Dialogic Inquiry (2026) Query Reformulation and Expansion 31 Question-to-Knowledge (Q2K): Multi-Agent Generation of Inspectable Facts for Product Mapping (2025) Entity Linking and Alignment |
|
| 23 | Keping Bi | 47 | 3 | 48 A Multi-Granularity-Aware Aspect Learning Model for Multi-Aspect Dense Retrieval (2024) Sparse and Dense Retrieval Encoders 47 CAME: Competitively Learning a Mixture-of-Experts Model for First-stage Retrieval (2025) Dense Bi-Encoder Training and Representation Learning |
|
| 24 | Weiwei Sun 0001 | 47 | 3 | 72 DukeNet: A Dual Knowledge Interaction Network for Knowledge-Grounded Conversation (2020) Conversational Search and Retrieval 65 Simulating User Satisfaction for the Evaluation of Task-oriented Dialogue Systems (2021) Conversational Search and Retrieval |
|
| 25 | Tetsuya Sakai | 46 | 3 | 86 ChatRetriever: Adapting Large Language Models for Generalized and Robust Conversational Dense Retrieval (2024) Conversational Search and Retrieval 59 Flexible Pseudo-Relevance Feedback Using Optimization Tables (2001) Query Reformulation and Expansion |
|
| 26 | Jing Liu 0022 | 46 | 3 | 50 RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking (2021) Dense Bi-Encoder Training and Representation Learning 47 Dense Text Retrieval Based on Pretrained Language Models: A Survey (2024) Dense Bi-Encoder Training and Representation Learning |
|
| 27 | Adam Jatowt | 45 | 3 | 86 Search Intent Discovery by Structurization of Community QA Contents (2012) Query Reformulation and Expansion 49 TempRetriever: Fusion-based Temporal Dense Passage Retrieval for Time-Sensitive Questions (2026) Domain-Specific and Structured Retrieval |
|
| 28 | Jinheon Baek | 45 | 3 | 70 Knowledge-Augmented Large Language Models for Personalized Contextual Query Suggestion (2024) Conversational Search and Retrieval 69 Realistic Conversational Question Answering with Answer Selection based on Calibrated Confidence and Uncertainty Measurement (2023) Conversational Search and Retrieval |
|
| 29 | Antonio Mallia | 45 | 3 | 59 Unveiling DIME: Reproducibility, Generalizability, and Formal Analysis of Dimension Importance Estimation for Dense Retrieval (2025) Query Reformulation and Expansion 44 Faster Learned Sparse Retrieval with Guided Traversal (2022) Learned Sparse Retrieval Efficiency |
|
| 30 | Yujia Zhou 0002 | 44 | 3 | 64 Cognitive Personalized Search Integrating Large Language Models with an Efficient Memory Mechanism (2024) Conversational Search and Retrieval 61 Incorporating Explicit Subtopics in Personalized Search (2023) User Behavior and Search Interaction |
|
| 31 | Venktesh V | 44 | 3 | 53 Understanding the User: An Intent-Based Ranking Dataset (2024) Query Reformulation and Expansion 51 The Surprising Effectiveness of Rankers trained on Expanded Queries (2024) Evidence Selection and Context Refinement |
|
| 32 | Chen Zhao 0013 | 44 | 3 | 51 Distantly-Supervised Dense Retrieval Enables Open-Domain Question Answering without Evidence Annotation (2021) Dense Bi-Encoder Training and Representation Learning 51 Multi-Step Reasoning Over Unstructured Text with Beam Dense Retrieval (2021) Multi-hop and Complex Reasoning QA |
|
| 33 | Zhe Xu 0009 | 43 | 3 | 72 Dynamic Demonstration Retrieval and Cognitive Understanding for Emotional Support Conversation (2024) Conversational Search and Retrieval 72 Intent-driven In-context Learning for Few-shot Dialogue State Tracking (2025) Conversational Search and Retrieval |
|
| 34 | Liwei Wang 0009 | 43 | 3 | 72 RAST: Domain-Robust Dialogue Rewriting as Sequence Tagging (2021) Conversational Search and Retrieval 65 Eliciting Knowledge from Large Pre-Trained Models for Unsupervised Knowledge-Grounded Conversation (2022) Knowledge Conflict and Faithfulness |
|
| 35 | Huiqiang Jiang | 43 | 3 | 72 SeCom: On Memory Construction and Retrieval for Personalized Conversational Agents (2025) Conversational Search and Retrieval 39 RetroInfer: A Vector Storage Engine for Scalable Long-Context LLM Inference (2026) Learned Sparse Retrieval Efficiency |
|
| 37 | Shu Zhou 0002 | 43 | 3 | 69 LOSDF: A logical optimization and semantic decoupling framework for question answering in multi-party conversations (2025) Conversational Search and Retrieval 69 Enhancing Extractive Question Answering in Multiparty Dialogues with Logical Inference Memory Network (2025) Conversational Search and Retrieval |