用大模型分析商品描述，揪出推荐系统里的刷单行为
SemanticShield: LLM-Powered Audits Expose Shilling Attacks in Recommender Systems
- 用大模型检测商品标题描述的语义异常，识别刷单
- 在六种攻击策略下检测准确率超90%，新攻击也有效
- 适合关注推荐系统安全的工程师和研究者
推荐系统广泛用于电商个性化推荐，但其开放性使其易受刷单攻击——攻击者通过注入虚假行为操纵推荐结果。现有防御多关注用户行为，忽略商品标题、描述等项侧特征暴露的恶意意图。为此，我们提出两阶段检测框架，通过大语言模型（LLM）整合项侧语义信息。第一阶段用低成本行为标准预筛选可疑用户，第二阶段采用基于LLM的审计机制评估语义一致性。进一步通过强化学习微调轻量级LLM，设计特定奖励函数，构建专用检测器SemanticShield。在六种典型攻击策略上验证其有效性，对未见过的攻击方法也展现强泛化能力。代码已开源：https://github.com/FrankenstLee/SemanticShield。
原文摘要 · Abstract (English)
Recommender systems (RS) are widely used in e-commerce for personalized suggestions, yet their openness makes them susceptible to shilling attacks, where adversaries inject fake behaviors to manipulate recommendations. Most existing defenses emphasize user-side behaviors while overlooking item-side features such as titles and descriptions that can expose malicious intent. To address this gap, we propose a two-stage detection framework that integrates item-side semantics via large language models (LLMs). The first stage pre-screens suspicious users using low-cost behavioral criteria, and the second stage employs LLM-based auditing to evaluate semantic consistency. Furthermore, we enhance the auditing model through reinforcement fine-tuning on a lightweight LLM with carefully designed reward functions, yielding a specialized detector called SemanticShield. Experiments on six representative attack strategies demonstrate the effectiveness of SemanticShield against shilling attacks, and further evaluation on previously unseen attack methods shows its strong generalization capability. Code is available at https://github.com/FrankenstLee/SemanticShield.
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。