Pre-search query recommendation, widely known as HintQ on Taobao's homepage, plays a vital role in intent capture and demand discovery, yet traditional methods suffer from shallow semantics, poor cold-start performance and low serendipity due to reliance on ID-based matching and co-click heuristics. To overcome these challenges, we propose AIGQ (AI-Generated Query architecture), the first end-to-end generative framework for HintQ scenario. AIGQ is built upon three core innovations spanning training paradigm, policy optimization and deployment architecture. First, we propose Interest-Aware List Supervised Fine-Tuning (IL-SFT), a list-level supervised learning approach that constructs training samples through session-aware behavior aggregation and interest-guided re-ranking strategy to faithfully model nuanced user intent. Accordingly, we design Interest-aware List Group Relative Policy Optimization (IL-GRPO), a novel policy gradient algorithm with a dual-component reward mechanism that jointly optimizes individual query relevance and global list properties, enhanced by a model-based reward from the online click-through rate (CTR) ranking model. To deploy under strict real-time and low-latency requirements, we further develop a hybrid offline-online architecture comprising AIGQ-Direct for nearline personalized user-to-query generation and AIGQ-Think, a reasoning-enhanced variant that produces trigger-to-query mappings to enrich interest diversity. Extensive offline evaluations and large-scale online A/B experiments on Taobao demonstrate that AIGQ consistently delivers substantial improvements in key business metrics across platform effectiveness and user engagement.
Figures & tables
Figure 1. Overview of HintQ
Figure 2. Overview of the AIGQ framework.
Model
Cate HR@30
Sem. Sim.
Query HR@30
Unique Cates
EBR
0.1998
0.5198
0.0100
8.4
Qwen3-30B-A3B
0.3054
0.5554
0.0022
5.4
Gemini 3 Pro
0.3449
0.5475
0.0017
7.0
GPT-5.1
0.3353
0.5606
0.0021
4.6
\textscAIGQ−DirectSFT
0.4181
0.5926
0.0428
7.1
\textscAIGQ−ThinkSFT
0.4437
0.6358
0.0549
7.7
Table 1. General performance on Taobao test set. The best results are illustrated in bold and the number underlined is the runner-up.
Model
Cate HR
Query HR
\textscAIGQ−DirectSFT
0.4181
0.0428
+ Interest-Guided Label Re-Ranking
0.4260
0.0442
+ Prompt Compression
0.4305
0.0442
+ IL-GRPO
0.4695
0.0679
\textscAIGQ−ThinkSFT
0.4437
0.0549
+ Interest-Guided Label Re-Ranking
0.4492
0.0594
Table 2. The experimental results of ablation study.
Category
Metric
Value (%)
Effectiveness
HintQ UCTR
+7.42
Attributed orders
+10.31
Attributed GMV
+10.68
Engagement
LT-7 retention
+3.73
Search UV
+0.20
Discovery
Unique queries
79.3
Table 3. Online performance of AIGQ across effectiveness, engagement, and discovery. Effectiveness and engagement report relative gains while discovery shows the percentage of queries or leaf categories uniquely covered by AIGQ .
Figure 3. Online A/B test: Impact of daily RL updates.
Query recommendation in e-commerce search aims to proactively suggest queries that match users' potential interests. However, existing methods mainly optimize query-level relevance, while neglecting whether the retrieved products align with users' downstream preferences. This mismatch often leads to high query click through rates (CTR) but low product conversion rates (CVR). To bridge this gap, we propose QueryAgent-R1, a memory-augmented agentic framework that improves end-to-end alignment via chain-of-retrieval optimization. Our QueryAgent-R1 grounds query generation in real inventory retrieval, allowing the agent to validate and refine queries based on retrieved products. We also design a consistency reward in the agentic reinforcement learning (RL) process to jointly optimize query relevance and downstream engagement. In addition, we construct a memory abstraction module for efficient user profiling. To support offline evaluation, we construct two datasets based on both proprietary industrial data and public datasets, on which QueryAgent-R1 consistently outperforms strong baselines. Moreover, on a large scale production platform, QueryAgent-R1 improves Query CTR by 2.9% and guided CVR by 3.1% in online A/B tests.
Generative query suggestion aims to enhance user engagement by anticipating user intents and recommending relevant follow-up queries. A central challenge is to generate slates whose individual queries are useful while the slate covers distinct intents. We propose an Intent-Driven Query Suggestion Framework with dual-stage optimization. First, intent-aware diversity modeling constructs intent-aligned supervised fine-tuning (SFT) data and uses an Intent-Aware Diversity Reward to optimize intent coverage. Second, query-level credit assignment routes individual quality signals to the corresponding query tokens while sharing a slate-level diversity signal across the slate. Experiments on a large-scale production dataset, including online A/B testing and offline evaluation, show improvements in click-through rate, query quality, and intent coverage.
Xinpeng Liu, Lu Ma, Jiayi Qiao +6
Peking University · Qwen Business Unit of Alibaba · National University of Singapore +1
Generative Retrieval (GR) has emerged as a promising paradigm by mapping queries directly to Semantic IDs (SIDs) with powerful representation capabilities for candidate items. However, existing SIDs derived solely from item content create a semantic gap, failing to align dynamic query intents with static item representations. Furthermore, current generative paradigms rarely model user behavior sequences and are always bottlenecked by the high inference latency of beam-search autoregressive decoding. To address these challenges, we propose Cross-component Hierarchical semantic Alignment for Personalized generative retrieval (CHAP), a novel personalized GR framework from a hierarchical perspective. First, we design a Hierarchical Semantic Alignment module to align query's latent space with item's quantization path and synchronize multi-granular semantics. Second, we construct a personalized GR framework that models user behavior by synergizing discrete SIDs for structural guidance and continuous representations for fine-grained semantic refinement. Notably, we introduce a Residual Cascading Generation mechanism to restrict the costly multi-step Transformer Decoder to a single-pass inference, boosting inference throughput while mitigating information loss. Extensive experiments on three public datasets, one proprietary industrial dataset, and online A/B tests demonstrate CHAP's superiority, validating the effectiveness and practical value of our approach. The code is publicly available at https://github.com/zzzgm/CHAP.
Gaoming Zhang, Angqing Jiang, Jianchun Song +4
University of Science and Technology of China · Meituan