2026-07-07 周二 · Personalization
独立报告页 · 在站内浏览（含往期切换）→Generative生成式推荐 · 3 篇Personalization个性化推荐 · 1 篇AI SearchAI 搜索 · 3 篇Agentic智能体搜索 · 10 篇Infra系统与索引 · 6 篇
今日场景社区搜索 (UGC 短文本搜索): 用户在社区平台搜索帖子、问答、评论等 UGC 内容。核心指标是搜索相关性 (nDCG) 和点击率 (CTR)。关注：语义匹配、query 理解、短文本召回、排序模型、LLM 在搜索中的应用。 — 评分 1 篇，入选 1 篇，分数 3.00 – 3.00
LEAD STORY · 今日必读
Fairness Attacks on Recommender Systems
这篇论文关注推荐系统中的公平性攻击问题，指出现有研究更多讨论推广/打压类攻击，对攻击者如何系统性放大推荐不公平性关注不足。作者提出结构感知的强化学习攻击框架，用图编码器建模伪造用户-物品交互与原始图结构的依赖，用循环网络刻画注入物品序列，并在目标模型使用性别等敏感属性时加入性别选择策略。实验覆盖四类目标推荐模型和两个真实数据集，结果显示该方法能有效加剧系统公平性偏差。
3.00
综合分 / 5
偏好调整 3.10
偏好调整 3.10
KEY INSIGHT · 对业务意味着什么对个性化推荐团队来说，这篇论文的价值不在于直接提升 CTR，而在于提醒线上排序系统需要把公平性鲁棒性纳入攻防评估。攻击策略同时操控结构、序列和敏感属性，说明仅靠常规异常用户过滤或单点鲁棒训练可能不足。实践中可借鉴其威胁建模方式，在召回、排序和重排离线评测中加入公平性压力测试，观察不同用户群体、物品群体的曝光和效用是否被少量伪造行为显著扭曲。
Yanan Wang, Yong Ge · 2026-06-27 · arXiv
其他赛道今日
点击切换开源雷达
GitHub / HuggingFace 热门 · 8 个项目colibri: Run GLM-5.2 (744B MoE) on a 25GB-RAM consumer machine — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦Run GLM-5.2 (744B MoE) on a 25GB-RAM consumer machine — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦C · ★ 47 · 2026-07-01 · GitHubglm-hammer: Forge-and-hammer engineering harness plugin for ZCode: 3-critic strong planning, worker-validator-critic implementation loop, evidence-gated hooksForge-and-hammer engineering harness plugin for ZCode: 3-critic strong planning, worker-validator-critic implementation loop, evidence-gated hooksHTML · ★ 5 · 2026-07-04 · GitHubGLM-5.2-655K-MTP-4x-DGX-Spark: Recipe: GLM-5.2 (unpruned QuantTrio Int4-Int8Mix) at 655,360-token context + MTP spec decode (DCP4, fp8_ds_mla) on a 4x NVIDIA DGX Spark (GB10) clusterRecipe: GLM-5.2 (unpruned QuantTrio Int4-Int8Mix) at 655,360-token context + MTP spec decode (DCP4, fp8_ds_mla) on a 4x NVIDIA DGX Spark (GB10) clusterShell · ★ 5 · 2026-07-05 · GitHubQwen/AgentWorldBenchAgentWorldBench AgentWorldBench is a comprehensive evaluation benchmark for language world models, constructed from real-world observations of frontier model trajectories on established benchmarks such as Tool Decathlon, Terminal-Bench 1.0 & 2.0, and OSWorld-Verified. Every evalu; tags: task_categories:text-generation, language:en, license:apache-2.0, size_categories:1K<n<10K, format:json, modality:tabular, modality:text, library:datasetsdataset · ★ 66 · 2026-06-22 · HuggingFacezai-org/GLM-5.2pipeline: text-generation; tags: transformers, safetensors, glm_moe_dsa, text-generation, conversational, en, zh, arxiv:2602.15763model · ★ 3520 · 2026-06-16 · HuggingFacedeepseek-ai/DeepSeek-V4-Pro-DSparkpipeline: text-generation; tags: transformers, safetensors, deepseek_v4, text-generation, arxiv:2606.19348, license:mit, endpoints_compatible, 8-bitmodel · ★ 407 · 2026-06-27 · HuggingFacezai-org/GLM-5.2-FP8pipeline: text-generation; tags: transformers, safetensors, glm_moe_dsa, text-generation, conversational, en, zh, arxiv:2602.15763model · ★ 188 · 2026-06-16 · HuggingFacedeepseek-ai/DeepSeek-V4-Flash-DSparkpipeline: text-generation; tags: transformers, safetensors, deepseek_v4, text-generation, arxiv:2606.19348, license:mit, endpoints_compatible, 8-bitmodel · ★ 166 · 2026-06-27 · HuggingFace