今日速览
| 序号 | 标题 | 来源 | 日期 | 主题 | 推荐等级 |
|---|---|---|---|---|---|
| 1 | Delegation Asymmetry in Agentic Recommender Systems: Measuring Two-Sided Receptivity in Online Dating | arXiv | 2026-08-18 | AI-Agent | 高 |
| 2 | StagedWorkspace: A Versioned Workspace for Knowledge-Work Agents | arXiv | 2026-08-18 | AI-Agent | 高 |
重点论文与技术动态
1. Delegation Asymmetry in Agentic Recommender Systems: Measuring Two-Sided Receptivity in Online Dating
- 来源：arXiv
- 日期：2026-08-18
- 作者/机构：Daria Leshchikova, Valentina V. Kuskova, Dmitry Zaytsev, Valerii Klimov
- 主题标签：
AI-Agent,arXiv - 推荐等级：高
- 分类：cs.AI
一句话结论
研究揭示在线约会平台中用户对代理委托存在系统性不对称，部署自身代理的接受度显著高于参与他人代理。
核心内容
- 用户发送（部署）与接收（参与）代理通信意愿高度相关（rho=0.92）但可分离，部署自身代理的接受度阈值（-0.38）远低于参与他人代理（+0.32）。
- 平均部署倾向是参与倾向的三倍，随机配对下仅4-13%的定向二元组能同时实现代理部署与接收参与，且存在性别方向不平衡。
- 互惠要求使交互量减少50%以上，而基于接收接受度路由可提升每联系参与度三倍（AUC 0.88，3.1x quartile lift）。
方法与数据
- 采用分级响应模型与潜在回归构建潜在变量测量模型。
- 基于两个大规模调查（N=2,894 和 N=2,617）收集数据，覆盖两种语言。
价值判断
- 值得关注：代理推荐系统需优先解决用户接受度不对称问题以提升匹配效率。
- 可复用点：设计反事实分析（如接收导向路由）可有效优化交互量与参与度。
- 局限/待核查：摘要未明确提及局限。
摘要
Autonomous LLM agents that converse on a user's behalf are an emerging design pattern in matching platforms, yet their viability depends on a condition rarely examined: users must accept not only delegating conversation to an agent, but also receiving agent-mediated communication from others. We study this condition using two large-scale surveys of active users of a major dating platform (N=2,894 on generative profile features; N=2,617 on autonomous conversational agents, fielded in two languages). We develop a latent-variable measurement model of agent receptivity based on graded response models with latent regression, and show via model comparison that willingness to send and willingness to receive agent communication are distinct constructs: highly correlated (rho=0.92) but separable (Delta BIC=52), with partial measurement invariance across languages. The model quantifies a systematic delegation asymmetry: deploying one's own agent requires far lower receptivity (threshold -0.38) than engaging a counterpart's agent (+0.32; full engagement +1.39), and mean deployment propensity exceeds engagement propensity roughly threefold. Under a random-pairing counterfactual derived from stated receptivity, only 4-13% of directed dyads combine agent deployment with receiver engagement, with a pronounced gender-directional imbalance. Design counterfactuals quantify the levers: a reciprocity requirement cuts interaction volume by half or more by excluding nearly two-thirds of would-be deployment, while routing agent contacts on receive receptivity triples per-contact engagement, a lift that survives out-of-sample validation with the target item held out (AUC 0.88, 3.1x quartile lift under respondent-level cross-validation). We discuss implications for agentic recommender design, including disclosure, opt-in mechanics, and receptivity-aware matchmaking.2. StagedWorkspace: A Versioned Workspace for Knowledge-Work Agents
- 来源：arXiv
- 日期：2026-08-18
- 作者/机构：Yining Hua, Hongbin Na, Yifan Zhou, Akshay Kalose, Cyrus Ayubcha, Levi Lian
- 主题标签：
AI-Agent,arXiv - 推荐等级：高
- 分类：cs.AI
一句话结论
StagedWorkspace通过显式绑定工作区状态到版本，显著提升知识工作代理的性能。
核心内容
- 知识工作代理处理代码库、文档等数字工件时，需显式绑定视图到工作区版本以避免版本混淆。
- StagedWorkspace使用内容哈希绑定解析记录和审查差异，确保状态一致性。
- 实验显示，双解析/原生访问比单视图提升OfficeQA Pass@1 8.3-12.1点和APEX平均评分4.7-9.2点；SW-AGENT在OfficeQA上得63.9%（对比29.3%），APEX上得42.1%（对比25.5%）。
方法与数据
- 基于OfficeQA Pro和APEX-Agents的固定框架消融实验。
- 摘要未明确具体数据集细节。
价值判断
- 值得关注：识别工作区状态为知识工作代理的实验变量，推动基于显式状态转换的基准测试。
- 可复用点：内容哈希绑定方法可推广至PDF、电子表格等多格式知识工作场景。
- 局限/待核查：摘要未明确局限。