用思维链让大模型更准判断电商搜索相关性
TaoSR1: The Thinking Model for E-commerce Relevance Search
- 引入思维链推理+多阶段优化，提升大模型判断能力
- 线上评测显示相关性评分显著优于现有方法
- 适合需要高精度搜索排序的电商平台使用
电商搜索中的查询-商品相关性预测是核心任务。基于BERT的模型在语义匹配上表现优异，但缺乏复杂推理能力；虽然大语言模型（LLMs）被探索用于此任务，多数仍采用判别式微调或蒸馏至小模型以部署。本文提出TaoSR1框架，直接部署LLM解决三大挑战：思维链（CoT）错误累积、判别式幻觉及部署可行性。该框架包含三个阶段：(1) 使用思维链进行监督微调（SFT），注入推理能力；(2) 离线采样结合pass@N策略与直接偏好优化（DPO），提升生成质量；(3) 基于难度的动态采样与组相对策略优化（GRPO），缓解判别式幻觉。此外，后思维链处理和基于累计概率的分块方法实现高效在线部署。TaoSR1在离线数据集上显著超越基线，在线上并行人类评估中取得显著提升，为思维链推理应用于相关性分类提供新范式。
原文摘要 · Abstract (English)
Query-product relevance prediction is a core task in e-commerce search. BERT-based models excel at semantic matching but lack complex reasoning capabilities. While Large Language Models (LLMs) are explored, most still use discriminative fine-tuning or distill to smaller models for deployment. We propose a framework to directly deploy LLMs for this task, addressing key challenges: Chain-of-Thought (CoT) error accumulation, discriminative hallucination, and deployment feasibility. Our framework, TaoSR1, involves three stages: (1) Supervised Fine-Tuning (SFT) with CoT to instill reasoning; (2) Offline sampling with a pass@N strategy and Direct Preference Optimization (DPO) to improve generation quality; and (3) Difficulty-based dynamic sampling with Group Relative Policy Optimization (GRPO) to mitigate discriminative hallucination. Additionally, post-CoT processing and a cumulative probability-based partitioning method enable efficient online deployment. TaoSR1 significantly outperforms baselines on offline datasets and achieves substantial gains in online side-by-side human evaluations, introducing a novel paradigm for applying CoT reasoning to relevance classification.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。