KDD

Less Traffic, Better Outcomes: Competition-Aware Request Dispatch in Real-Time Ad Exchanges

更少流量，更好结果：实时广告交易平台中感知竞争的请求调度

Jonaid Shianifar, Blaz Mramor, Fangda Zou, Matthieu C. Martin, Xingsheng Guo, Zhihua Zhu, Rong Zhou, Bichen Shi

AI总结 该研究针对实时广告交易平台请求过度分发的问题，提出感知竞争的请求调度框架，经在线实验验证可降低DSP请求量并提升净收入。

Comments Accepted for presentation at AdKDD 2026, the premier workshop on artificial intelligence for advertising, held in conjunction with the 32nd ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2026)

Traceable Multi-Agent System for Knowledge-Based Forecasting

用于基于知识的预测的可追踪多智能体系统

Junhyeok Kang, Sangjun Han, Hyeokjun Choe, Soonyoung Lee

AI总结 本文提出可追踪多智能体预测演示系统TraceMAS，通过两类因果环图关联证据与预测，在原油价格预测中验证其可让自主智能体兼顾灵活性与过程可检查性。

Comments Accepted at KDD 2026 Workshop on Enterprise AI Agents (Oral)

FakeI2V-Bench: Benchmarking the Applicability of Image-level Deepfake Detectors for Deepfake Video Detection

FakeI2V-Bench：评估图像级深度伪造检测器在深度伪造视频检测中的适用性

Pei Li, Sihan Chen, Delong Ran, Tianshuo Cong

AI总结 本研究构建FakeI2V-Bench基准，评估图像级深度伪造检测器在视频检测中的性能，提出IV-Bridge框架聚合帧级预测，使多数图像级检测器性能超越现有视频级方法，为相关研究提供基准与新方向。

Comments To Appear in KDD 2026, Jeju, Korea, August 9-13, 2026

GeoID-PINN: Identifiability-Aware Regional Epidemic Inference with Geographic Coupling

GeoID-PINN：考虑地理耦合的可识别性感知区域流行病推断方法

Weixiong Hua, Fan Bu

机构 * University of Michigan（密歇根大学） 纠错

AI总结 本研究提出GeoID-PINN，一种结合地理耦合与正则化的物理信息神经网络，用于区域流行病SIRD动力学推断，在模拟数据与路易斯安那州64个县COVID-19数据上均提升了预测准确性。

Comments 11 pages, 3 figures. Accepted at the 8th epiDAMIK ACM SIGKDD Workshop on Data-driven Decision Making for Public and Population Health (epiDAMIK @ KDD 2026)