Homepage - Weilin Ruan

Weilin Ruan (阮炜霖)

Incoming PhD Student @ CUHK CSE

MPhil Student @ HKUST(GZ)

Bachelor of Engineering @ Jinan University

I am an incoming PhD student in Computer Science and Engineering at The Chinese University of Hong Kong (CUHK), supervised by Prof. James Cheng, jointly trained with Knowin AI Inc. (Shenzhen). I received my MPhil from HKUST(GZ) supervised by Prof. Yuxuan Liang.

My research interests include Spatio-Temporal World/Action Models, Embodied Foundation Models, Data Generation & Verification, and Infrastructure (Infra) for AI systems. And my research experiences include Spatio-Temporal Data Mining, Time Series Forecasting, and Multi-Modal Learning.

Milestones

Key progress in research and academic activities.

Publications 15+
Citations 100+
Awards 20+
Blog Fans 3600+

Interests

Beyond research, I enjoy these activities in daily life.

Video Games Badminton Hiking Movies Cars Travel Photography Music Indie Game Dev Pets Chess Board Games Web Novels

View More Hobbies

Education

The Chinese University of Hong Kong - Computer Science and Engineering
Supervised by Prof. James Cheng | Joint training with Knowin AI

Aug 2026 -
HKUST(GZ) - Information Hub - Data Science and Analysis

Sep 2024 - June 2026
Jinan University - Computer Science Department - Network Engineering

Sep 2020 - June 2024

Latest News

Paper Acceptance - One paper accepted to KDD 2026.

May 2026
New Preprint - MAS4TS: Visual Reasoning over Time Series via Multi-Agent System is available on arXiv (under review).

Feb 2026
Citation Milestone - Google Scholar citations reached 100+.

Mar 2026
Paper Acceptance - One paper accepted to ICLR Workshop HCAIR 2026.

Mar 2026
Paper Acceptance - Two papers accepted to AAAI 2026.

Nov 2025
Reviewer Appointment - Appointed as a reviewer for ICLR 2026 conference.

Sep 2025
New Position - Joined Knowin AI Inc. as an Algorithm Researcher.

Sep 2025
Program Committee - Invited to serve as a Program Committee member for AAAI 2026.

July 2025
Paper Acceptance - Paper accepted to ICCV 2025 Workshop.

July 2025
Paper Acceptance - Paper on Traffic Flow Forecasting accepted to IEEE TITS Journal.

Aug 2025
Paper Acceptance - Paper on Multimodal Building Electricity Loads Forecasting accepted to ACM MM 2025.

July 2025
Paper Acceptance - Paper on Urban Heat Island Effect Forecasting accepted to KDD 2025.

May 2025
Paper Acceptance - Paper on Spatio-Temporal Forecasting accepted to ECML-PKDD 2025.

May 2025
Paper Acceptance - Paper on Multimodal Time Series Forecasting accepted to ICML 2025.

May 2025
Reviewer Appointment - Appointed as a reviewer for IJCNN and ICASSP conferences.

Dec 2024
Program Committee - Invited to serve as a Program Committee member for WebST 2025.

Jan 2025
Outstanding Graduate - Received the undergraduate certificate from Jinan University as an outstanding graduate of the department of Information Science and Technology.

June 2024
Research Internship - Joined the CityMind team as an intern led by Prof. Yuxuan Liang.

Aug 2023
RBCC Achievement - Participated in the RBCC offline held in HKUST(GZ) and achieved outstanding camper.

July 2023

Pinned Publications

TS-Memory: Plug-and-Play Memory for Time Series Foundation Models Pinned

Sisuo Lyu, Siru Zhong, Tiegang Chen, Weilin Ruan, Qingxiang Liu, Taiqiang Lv, Qingsong Wen, Raymond Chi-Wing Wong, Yuxuan Liang

KDD. 2026

We propose TS-Memory, a lightweight plug-and-play memory adapter that augments frozen Time Series Foundation Models. TS-Memory distills offline retrieval's distributional knowledge into a compact neural module via Parametric Memory Distillation, enabling retrieval-free deployment with constant-time inference while achieving state-of-the-art performance across diverse benchmarks.

Paper

Visual Reasoning over Time Series via Multi-Agent System Pinned Under Review

Weilin Ruan, Yuxuan Liang

Under Review. 2026

We propose MAS4TS, a tool-driven multi-agent framework for general time-series tasks under an Analyzer-Reasoner-Executor paradigm. It integrates visual reasoning over time-series plots, latent trajectory reconstruction, and gated inter-agent communication to improve cross-task generalization and inference efficiency.

Paper

OccamVTS: Distilling Vision Models to 1% Parameters for Time Series Forecasting Pinned

Sisuo Lyu, Siru Zhong, Weilin Ruan, Qingxiang Liu, Qingsong Wen, Hui Xiong, Yuxuan Liang

AAAI. 2026

We propose OccamVTS, a novel framework that distills large vision models to only 1% of their original parameters for efficient time series forecasting, demonstrating that extreme parameter reduction can be achieved while maintaining strong predictive performance through innovative distillation techniques.

Paper Code

Retrieval Augmented Spatio-Temporal Framework for Traffic Prediction Pinned

Weilin Ruan, Xilin Dang, Ziyu Zhou, Sisuo Lyu, Yuxuan Liang

AAAI. 2026

We propose RAST, a universal framework that integrates retrieval-augmented mechanisms with spatio-temporal modeling to address limited contextual capacity and low predictability in traffic prediction. Our framework consists of three key designs: Decoupled Encoder and Query Generator, Spatio-temporal Retrieval Store and Retrievers, and Universal Backbone Predictor that flexibly accommodates pre-trained STGNNs or simple MLP predictors.