paper.dinq.meAI 与前沿科学顶会 · 学术新星雷达
登录 DINQ
返回新星榜
TA

Tmall Group of Alibaba)

AAAI 2026IROS 2026SIGIR 2026效率与系统多模态评测与基准对齐与偏好可解释性
新星分新星
71
一作贡献30
核心/参与10
方向广度16
合作网络12
1 一作 · 3 篇 · AAAI 2026 / IROS 2026 / SIGIR 2026

获取完整研究档案

首次获取该候选人的完整档案及 StartLux 评估共 20 积分;每位用户单独解锁,之后查看免费。主动刷新 20 积分。

联系方式

实时查找电子邮箱,费用不超过20积分;未找到邮箱不扣费。

2026 顶会论文

3
作者位次 #6/11SIGIR 2026Poster· Information Retrieval (Industry)

PGSF: Profile-Guided Semantic Fusion for Decoupled Industrial Pre-Ranking

Chuike Sun (Taobao &amp · Tmall Group of Alibaba) · Tmall Group of Alibaba) · Tmall Group of Alibaba) · Tmall Group of Alibaba) · Tmall Group of Alibaba) · Tmall Group of Alibaba) · Shengjie Sun (Taobao &amp · Tmall Group of Alibaba) · Shengjie Sun (Taobao &amp · Tmall Group of Alibaba)

摘要

Pre-ranking is a critical stage in large-scale recommendation systems, strictly demanding both high efficiency and effectiveness. Despite being the industry benchmark for pre-ranking owing to its serving efficiency, the two-tower structure, by virtue of its decoupled architecture, inherently rest...

效率与系统评测与基准对齐与偏好OpenReview
一作IROS 2026Poster· Putting Language Models in the Control Loop

Fast Enough to Act: Spatio-Temporal Visual Token Merging for Low-Latency Robotic VLMs and VLAs

Tmall Group of Alibaba) · Jindong Wang · Gang Zhou

多模态效率与系统Optimization and Optimal ControlAI-Enabled RoboticsAI-Based MethodsOpenReview

常见合作者