Modality-Decoupled Online Recursive Editing
Siyuan Li, Youyuan Zhang, Fangming Liu, Jing Li
ICML 2026
A modality-decoupled approach to online recursive model editing for MLLMs.
Publications
A collection of my research work.
Evolving Beyond Snapshots: Harmonizing Structure and Sequence via Entity State Tuning for Temporal Knowledge Graph Forecasting
Siyuan Li, Yunjia Wu, Yiyong Xiao, Pingyang Huang, Peize Li, Ruitong Liu, Yan Wen, Te Sun
ACL Main Conference 2026
An entity-state tuning framework that harmonizes structural and sequential signals for long-horizon temporal knowledge graph forecasting.
Toward Robust Personalized Alignment for LLMs: Mitigating Persona Drift in Multi-Turn Dialogue
Youyuan Zhang, Siyuan Li, Fangming Liu, Jing Li
EMNLP 2026 Findings 2026
A robust personalized alignment approach for mitigating persona drift in multi-turn dialogue.
Multi-Condition Latent Diffusion Network for Semantic-Aware Knowledge Graph Completion
Siyuan Li, Liang Zhang, Bo Jin, Xiaopeng Wei
Knowledge-Based Systems 2026
A multi-condition latent diffusion network for semantic-aware knowledge graph completion.
From Decoration to Co-Presence: Reconfiguring Human--Plant Relations in Urban Workplaces Through Aesthetic Bio-Feedback
Yibo Meng, Bingyi Liu, Siyuan Li, Xinran Huang
Design Research Society International Conference 2026
A research-through-design study exploring how aesthetic bio-feedback can reconfigure human--plant relations in urban workplaces.
Mouse Heartbeat: Exploring Perceptual Augmentation for Ethical Awareness
Yibo Meng, Bingyi Liu, Siyuan Li, Yan Guan
Augmented Humans International Conference 2026
An exploration of perceptual augmentation as a means of supporting ethical awareness through heartbeat interaction.
Heating Load Profiles Analysis and Probabilistic Prediction for a Public Building: A Data-Driven Deep Learning Approach
Yi Xu, Siyuan Li, Xiangli Li, Meng Gao, Weiqiang Kong, Jianhua Fan, Haocheng Zhang
Journal of Building Engineering 2025
A data-driven deep learning approach for analyzing heating load profiles and probabilistic prediction in a public building.
Multimodal Tactile Feedback System for Ankle and Knee Protection during Physical Exercise: A Ubiquitous Computing Approach
Yibo Meng, Zhiming Liu, Siyuan Li, Te Sun
Proceedings of the 2025 ACM International Joint Conference on Pervasive and Ubiquitous Computing and the 2025 ACM International Symposium on Wearable Computing Companion 2025
A multimodal tactile feedback system designed to support ankle and knee protection during physical exercise.
Flow-Modulated Scoring for Semantic-Aware Knowledge Graph Completion
Siyuan Li, Ruitong Liu, Yan Wen, Te Sun, Andi Zhang, Yanbiao Ma, Xiaoshuai Hao
arXiv preprint arXiv:2506.23137 2025
A flow-modulated scoring framework that combines semantic context learning with conditional flow matching for knowledge graph completion.
Context-Driven Knowledge Graph Completion with Semantic-Aware Relational Message Passing
Siyuan Li, Ruitong Liu, Te Sun, Yan Wen, Ruihao Zhou, Jingyi Kang, Yunjia Wu
International Conference on Neural Information Processing 2025
A semantic-aware relational message-passing framework that selects relevant graph context for robust knowledge graph completion.
Beyond Prefixes: Graph-as-Memory Cross-Attention for Knowledge Graph Completion with Large Language Models
Ruitong Liu, Boxu Lin, Peize Li, Yunjia Wu, Te Sun, Chaohan Wu, Siyuan Li
arXiv preprint arXiv:2510.08966 2025
A graph-as-memory cross-attention framework that moves beyond prefix-based knowledge injection for knowledge graph completion with large language models.
Event USKT: U-State Space Model in Knowledge Transfer for Event Cameras
Yuhui Lin, Jiahao Zhang, Siyuan Li, Jimin Xiao, Ding Xu, Wenjun Wu, Jiaxuan Lu
arXiv preprint arXiv:2411.15276 2024
A U-shaped state space model framework for transferring knowledge from RGB models to event-camera data with minimal parameter tuning.