Research map / Label-efficient and robust learning / Recognition and long-tail
OneSug: The Unified End-to-End Generative Framework for E-commerce Query Suggestion
Xian Guo, Ben Chen, Siyuan Wang, Ying Yang, Mingyue Cheng, Chenyi Lei, Yuqing Ding, Han Li
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace multi-stage cascade with end-to-end generative OneSug (BART-base) | e-commerce query suggestion · Kuaishou test (Click) | HR@16: 78.61% -> 82.14% | Improved |
| scale decoder-only base model from 0.5B to 3B (ablation) | OneSug Qwen2.5 · Kuaishou test (Click) | HR@16: 85.58% -> 93.37% | Improved |
| use bigger encoder-decoder (mT5 small to base) (ablation) | OneSug · Kuaishou test (Click) | HR@16: 82.01% -> 83.63% | Improved |
| add DPO preference alignment to SFT generative model (ablation) | GRA · Kuaishou test (Click) | HR@16: 73.16% -> 75.50% | Improved |
| add reward-weighted ranking (ablation) | OneSug Pair BART-B · Kuaishou test (Click) | HR@16: 77.28% -> 79.39% | Improved |
| add margin loss (ablation) | OneSug Pair BART-B · Kuaishou test (Click) | HR@16: 78.81% -> 79.39% | Improved |
| add reward-gap weighting (ablation) | OneSug Pair BART-B · Kuaishou test (Click) | MRR: 44.41% -> 46.89% | Improved |
| switch from pair-wise to list-wise DPO (ablation) | OneSug BART-B · Kuaishou test (Click) | HR@16: 79.39% -> 82.14% | Improved |
| switch from pair-wise to list-wise DPO (ablation) | OneSug online · online A/B | CTR: +1.78% -> +2.01% | Improved |
| add prefix2query representation enhancement (ablation) | OneSug · Kuaishou test | HR@16 | Improved |
| lengthen related query sequence (ablation) | OneSug inference · Kuaishou test | HR@16 | Mixed |
| increase beam size from 16 to 128 (ablation) | OneSug inference · Kuaishou test | HR@16 | Improved |
Closest papers by meaning
- OneSearch-V2: The Latent Reasoning Enhanced Self-distillation Generative Search Framework (NeurIPS 2026 · similarity 0.89)
- Sparse Meets Dense: Unified Generative Recommendations with Cascaded Sparse-Dense Representations (NeurIPS 2025 · similarity 0.82)
- eCeLLM: Generalizing Large Language Models for E-commerce from Large-scale, High-quality Instruction Data (ICML 2024 · similarity 0.82)
- EcoGEO: Trajectory-Aware Evidence Ecosystems for Web-Enabled LLM Search Agents (NeurIPS 2026 · similarity 0.82)
- EcomGPT: Instruction-Tuning Large Language Models with Chain-of-Task Tasks for E-commerce (AAAI 2024 · similarity 0.81)
- LLM-ESR: Large Language Models Enhancement for Long-tailed Sequential Recommendation (NeurIPS 2024 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
