Yuhang Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yuhang Wang has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Chuyu Wang
College of Computer and Information Science, Southwest University, Chongqing, China
1 shared publication - Dan Zhao
Pengcheng Laboratory, Shenzhen, China
1 shared publication - Jingyu Xiao
Chinese University of Hong Kong, Hong Kong, Hong Kong
1 shared publication - Mingjie Wang
Department of Mathematics, Zhejiang Sci-Tech University, Hangzhou, China
1 shared publication - Qing Li
Pengcheng Laboratory, Shenzhen, China
1 shared publication - Qingsong Zou
Tsinghua University
1 shared publication
Latest Publications
- Open Access
- 20 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1751-1762
SmartGen: Synthesizing Context-Aware User Behavior Data for Adaptive Smart Home Intelligence
As smart homes become increasingly prevalent, intelligent models are widely used for tasks such as anomaly detection and behavior prediction. These models are typically trained on static datasets, making them brittle to behavioral drift caused by seasonal changes, lifestyle shifts, or evolving routines. However, collecting new behavior data for retraining is often impractical due to its slow pace, high cost, and privacy concerns. In this paper, we propose SmartGen, an LLM-based framework that synthesizes context-aware user behavior data to support continual adaptation of downstream smart home models. SmartGen consists of four key components. First, we design a Time and Semantic-aware Split module to divide long behavior sequences into manageable, semantically coherent subsequences under dual time-span constraints. Second, we propose Semantic-aware Sequence Compression to reduce input length while preserving representative semantics by clustering behavior mapping in latent space. Third, we introduce Graph-guided Sequence Synthesis, which constructs a behavior relationship graph and encodes frequent transitions into prompts, guiding the LLM to generate data aligned with contextual changes while retaining core behavior patterns. Finally, we design a Two-stage Outlier Filter to identify and remove implausible or semantically inconsistent outputs, aiming to improve the factual coherence and behavioral validity of the generated sequences. Experiments on three real-world datasets demonstrate that SmartGen significantly enhances model performance on anomaly detection and behavior prediction tasks under behavioral drift, with anomaly detection improving by 85.43% and behavior prediction by 70.51% on average. The code is available at https://github.com/xzyvoid/SmartGen.
- 20 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 1751-1762
Open AccessSmartGen: Synthesizing Context-Aware User Behavior Data for Adaptive Smart Home Intelligence
As smart homes become increasingly prevalent, intelligent models are widely used for tasks such as anomaly detection and behavior prediction. These models are typically trained on static datasets, making them brittle to behavioral drift caused by seasonal changes, lifestyle shifts, or evolving routines. However, collecting new behavior data for retraining is often impractical due to its slow pace, high cost, and privacy concerns. In this paper, we propose SmartGen, an LLM-based framework that synthesizes context-aware user behavior data to support continual adaptation of downstream smart home models. SmartGen consists of four key components. First, we design a Time and Semantic-aware Split module to divide long behavior sequences into manageable, semantically coherent subsequences under dual time-span constraints. Second, we propose Semantic-aware Sequence Compression to reduce input length while preserving representative semantics by clustering behavior mapping in latent space. Third, we introduce Graph-guided Sequence Synthesis, which constructs a behavior relationship graph and encodes frequent transitions into prompts, guiding the LLM to generate data aligned with contextual changes while retaining core behavior patterns. Finally, we design a Two-stage Outlier Filter to identify and remove implausible or semantically inconsistent outputs, aiming to improve the factual coherence and behavioral validity of the generated sequences. Experiments on three real-world datasets demonstrate that SmartGen significantly enhances model performance on anomaly detection and behavior prediction tasks under behavioral drift, with anomaly detection improving by 85.43% and behavior prediction by 70.51% on average. The code is available at https://github.com/xzyvoid/SmartGen.
- Open Access
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 416-423
Open-World 3D Scene Understanding with Cross-Modal Dual Consistency Learning
- 30 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 416-423
Open AccessOpen-World 3D Scene Understanding with Cross-Modal Dual Consistency Learning