Demand-Calibrated Facet Diversification with Deployable Soft Quotas for E-Commerce Search
sigir26-114 · Full research paper · Shenghui Xu, Jiang Yu
User Behavior and Search Interaction · 50% Personalized and Context-Aware Retri · 25% Evaluation Metrics and User Models · 25%
Abstract
E-commerce search rankers optimized for engagement or conversion often homogenize first-page results, under-serving ambiguous or exploratory queries. Prior diversification methods typically assume per-query intent/aspect supervision or heavy re-ranking, which is costly and brittle under production constraints. We proposed a supervision-free, deployable framework for demand-calibrated facet diversification. From independent holdout traffic, we estimate a cohort-level target facet distribution using session-attributed downstream engagement, capturing how much diversity users actually demand. At serving time, we steer the first-page slate toward this target with a lightweight soft-quota score adjustment over a short candidate prefix, preserving the base ranker's within-facet ordering and adding negligible latency. Large-scale online experiments on high-volume traffic show consistent improvements in revenue and engagement while reducing abandonment. We further find that delivered first-page facet mixes move closer to the engagement-derived target, indicating effectiveness of the proposed mechanism.
Assigned reviewers
| # | Reviewer | Match | Why (closest evidence publications in the taxonomy) | |
|---|---|---|---|---|
| 1 | Jaime Arguello | 69 | 86 The Effects of Vertical Rank and Border on Aggregated Search Coherence and Search Behavior (2014) User Behavior and Search Interaction 86 The effect of aggregated search coherence on search behavior (2012) User Behavior and Search Interaction 71 The effects of goal-setting on learning during information seeking with generative AI (2026) User Behavior and Search Interaction |
|
| 2 | Mark D. Smucker | 58 | 86 Time-based calibration of effectiveness measures (2012) User Behavior and Search Interaction 86 Evaluating Streams of Evolving News Events (2015) User Behavior and Search Interaction 86 IR Evaluation: Modeling User Behavior for Measuring Effectiveness (2015) User Behavior and Search Interaction |
|
| 3 | Xuanhui Wang | 51 | 70 Revisiting Two-tower Models for Unbiased Learning to Rank (2022) User Behavior and Search Interaction 70 Learning with Sparse and Biased Feedback for Personal Search (2018) User Behavior and Search Interaction 67 Cross-Positional Attention for Debiasing Clicks (2021) User Behavior and Search Interaction |
Area chair
| Area chair | Match | |
|---|---|---|
| Ziyi Ye | 55 |
Area-chair candidates (6)
| Ziyi Ye assigned | 55 | 14 papers | |
| Fernando Diaz 0001 | 53 | 12 papers | |
| Zhijing Wu 0001 | 45 | 9 papers | |
| Jun Xu 0001 | 38 | 2 papers | |
| Chenyan Xiong | 35 | 15 papers | |
| Xiaoxue Zang | 34 | 4 papers |
Candidates from the panel ranked by match score
| # | Reviewer | Match | Load | Why | |
|---|---|---|---|---|---|
| 1 | Vuong Thanh Tung | 78 | 3 | 87 Does More Context Help? Effects of Context Window and Application Source on Retrieval Performance (2022) User Behavior and Search Interaction 86 Predicting Representations of Information Needs from Digital Activity Context (2024) User Behavior and Search Interaction |
|
| 2 | Fan Zhang 0053 | 71 | 3 | 88 Incorporating Query Reformulating Behavior into Web Search Evaluation (2021) User Behavior and Search Interaction 87 Towards Better Evaluating Multi-Query Sessions: A Measure Based on the Theory of Planned Behavior (2025) User Behavior and Search Interaction |
|
| 3 | Sara Allawati | 70 | 3 | 75 A Comparative Analysis of Linguistic and Retrieval Diversity in LLM-Generated Search Queries (2025) User Behavior and Search Interaction 67 What do Users Really Ask Large Language Models? An Initial Log Analysis of Google Bard Interactions in the Wild (2024) User Behavior and Search Interaction |
|
| 4 | Jiqun Liu | 69 | 3 | 67 Matching Search Result Diversity with User Diversity Acceptance in Web Search Sessions (2022) User Behavior and Search Interaction 67 Constructing and meta-evaluating state-aware evaluation metrics for interactive search systems (2023) User Behavior and Search Interaction |
|
| 5 | Jaime Arguello assigned | 69 | 3 | 86 The Effects of Vertical Rank and Border on Aggregated Search Coherence and Search Behavior (2014) User Behavior and Search Interaction 86 The effect of aggregated search coherence on search behavior (2012) User Behavior and Search Interaction |
|
| 6 | Mark D. Smucker assigned | 58 | 3 | 86 Time-based calibration of effectiveness measures (2012) User Behavior and Search Interaction 86 Evaluating Streams of Evolving News Events (2015) User Behavior and Search Interaction |
|
| 9 | Xuanhui Wang assigned | 51 | 3 | 70 Revisiting Two-tower Models for Unbiased Learning to Rank (2022) User Behavior and Search Interaction 70 Learning with Sparse and Biased Feedback for Personal Search (2018) User Behavior and Search Interaction |
|
| 10 | Tetsuya Sakai | 51 | 3 | 54 Diversification as Risk Minimization (2026) Evaluation Metrics and User Models 54 Evaluation with informational and navigational intents (2012) Evaluation Metrics and User Models |
|
| 11 | Falk Scholer | 51 | 3 | 75 Classifying Term Variants in Query Formulation (2025) User Behavior and Search Interaction 75 Can Users Predict Relative Query Effectiveness? (2022) User Behavior and Search Interaction |
|
| 12 | Romain Deffayet | 49 | 3 | 67 An Offline Metric for the Debiasedness of Click Models (2023) User Behavior and Search Interaction 67 Evaluating the Robustness of Click Models to Policy Distributional Shift (2023) User Behavior and Search Interaction |
|
| 13 | Leif Azzopardi | 49 | 3 | 86 User Models, Metrics and Measures of Search: A Tutorial on the C/W/L Evaluation Framework (2021) User Behavior and Search Interaction 75 Query quality: user ratings and system predictions (2010) User Behavior and Search Interaction |
|
| 14 | Paul Thomas 0001 | 48 | 3 | 86 User Models, Metrics and Measures of Search: A Tutorial on the C/W/L Evaluation Framework (2021) User Behavior and Search Interaction 71 Large Language Models can Accurately Predict Searcher Preferences (2024) User Behavior and Search Interaction |
|
| 16 | Saber Zerhoudi | 42 | 3 | 71 Beyond the Click: A Framework for Inferring Cognitive Traces in Search (2026) User Behavior and Search Interaction 71 In-Browser Agents for Search Assistance (2026) User Behavior and Search Interaction |
|
| 17 | Shuaiqiang Wang | 40 | 3 | 74 FULTR: A Large-Scale Fusion Learning to Rank Dataset and Its Application for Satisfaction-Oriented Ranking (2025) User Behavior and Search Interaction 72 M2oERank: Multi-Objective Mixture-of-Experts Enhanced Ranking for Satisfaction-Oriented Web Search (2025) User Behavior and Search Interaction |
|
| 18 | Zhen Qin 0001 | 39 | 3 | 70 Revisiting Two-tower Models for Unbiased Learning to Rank (2022) User Behavior and Search Interaction 70 Cross-Positional Attention for Debiasing Clicks (2021) User Behavior and Search Interaction |
|
| 19 | Idan Szpektor | 39 | 3 | 67 Predicting web searcher satisfaction with existing community-based answers (2011) User Behavior and Search Interaction 53 Identifying Web Queries with Question Intent (2016) Personalized and Context-Aware Retrieval |
|
| 21 | Keping Bi | 37 | 3 | 71 A Study of Context Dependencies in Multi-page Product Search (2019) User Behavior and Search Interaction 47 Learning a Fine-Grained Review-based Transformer Model for Personalized Product Search (2021) Recommendation with Search and Ranking |
|
| 22 | Gustavo Penha | 37 | 3 | 71 As It Was: Aligning LLM Search Evaluation with Historical User Preferences (2026) User Behavior and Search Interaction 46 Pairwise Review-Based Explanations for Voice Product Search (2022) Conversational Search and Retrieval |
|
| 23 | Yujia Zhou 0002 | 37 | 3 | 69 Intent-Oriented Dynamic Interest Modeling for Personalized Web Search (2024) User Behavior and Search Interaction 69 Incorporating Explicit Subtopics in Personalized Search (2023) User Behavior and Search Interaction |
|
| 24 | Daiting Shi | 37 | 3 | 54 MA4DIV: Multi-Agent Reinforcement Learning for Search Result Diversification (2025) Evaluation Metrics and User Models 52 Accurate and Efficient Personalized Query Rewriting in Baidu Search (2026) Query Reformulation and Expansion |
|
| 25 | Xiao Zhang 0034 | 37 | 3 | 48 Similarity = Value? Consultation Value-Assessment and Alignment for Personalized Search (2025) Personalized and Context-Aware Retrieval 37 UniSAR: Modeling User Transition Behaviors between Search and Recommendation (2024) Recommendation with Search and Ranking |
|
| 26 | Negar Arabzadeh | 36 | 3 | 41 Query Performance Prediction: From Ad-hoc to Conversational Search (2023) Query Performance Prediction 39 Matches Made in Heaven: Toolkit and Large-Scale Datasets for Supervised Query Reformulation (2021) Query Reformulation and Expansion |
|
| 28 | Yang Song 0008 | 34 | 3 | 73 Query-Less: Predicting Task Repetition for NextGen Proactive Search and Recommendation Engines (2016) User Behavior and Search Interaction 37 GenSAR: Unifying Balanced Search and Recommendation with Generative Retrieval (2025) Recommendation with Search and Ranking |
|
| 29 | Bo Peng 0009 | 34 | 3 | 68 SWaT: Statistical Modeling of Video Watch Time through User Behavior Analysis (2025) User Behavior and Search Interaction 32 Hybrid collaborative filtering methods for recommending search terms to clinicians (2021) Recommendation with Search and Ranking |
|
| 31 | Rolf Jagerman | 33 | 3 | 83 Improving Cloud Storage Search with User Activity (2021) User Behavior and Search Interaction 37 Bootstrapping Recommendations at Chrome Web Store (2021) Recommendation Systems and Evaluation |
|
| 32 | Zechun Niu | 33 | 3 | 36 Investigating the Robustness of Counterfactual Learning to Rank Models: A Reproducibility Study (2025) Unbiased Learning to Rank 32 Addressing Personalized Bias for Unbiased Learning to Rank (2025) Unbiased Learning to Rank |
|
| 33 | Yifan Wang 0017 | 33 | 3 | 71 User Perception of Recommendation Explanation: Are Your Explanations What Users Need? (2023) User Behavior and Search Interaction 31 Improving Long-tail User CTR Prediction via Hierarchical Distribution Alignment (2025) Bias Mitigation Methods |
|
| 34 | Tingyu Song | 33 | 3 | 71 IFIR: A Comprehensive Benchmark for Evaluating Instruction-Following in Expert-Domain Information Retrieval (2025) User Behavior and Search Interaction 31 Rethinking Reasoning-Intensive Retrieval: Evaluating and Advancing Retrievers in Agentic Search Systems (2026) Reasoning-Intensive Retrieval and Ranking |
|
| 35 | Sunhao Dai | 33 | 3 | 71 NExT-Search: Rebuilding User Feedback Ecosystem for Generative AI Search (2025) User Behavior and Search Interaction 32 Uncovering ChatGPT's Capabilities in Recommender Systems (2023) Recommendation with Search and Ranking |
|
| 36 | Jialong Wu 0007 | 33 | 3 | 36 AdaRewriter: Unleashing the Power of Prompting-based Conversational Query Reformulation via Test-Time Adaptation (2025) Query Reformulation and Expansion 36 AdaCQR: Enhancing Query Reformulation for Conversational Search via Sparse and Dense Retrieval Alignment (2025) Query Reformulation and Expansion |