I am a Machine Learning Quant Algorithm Engineer in the Quantitative Technology Department at X-tech, working on machine learning algorithms and quantitative strategy research.

I received my B.S. from the School of Information Science and Engineering, Chongxin College, Shandong University, where I was supervised by Prof. Zhi Liu. Before joining X-tech, I worked on quantitative research and AI algorithms at JoinQuant, served as a Research Assistant at HKUST(GZ) AI Thrust & INFO Hub under Prof. Yutao Yue, and conducted trustworthy AI research as a remote research intern at KAUST.

My research centers on trustworthy AI, especially concept bottleneck models (CBMs). I study editable and continual CBMs, multimodal interpretability, concept-level security, reliable time-series forecasting, medical AI, autonomous driving, and LLM interpretability. My work has been published or accepted at venues including ICLR, NeurIPS, ACM MM, ECCV, MICCAI, EMNLP, TMLR, ECML-PKDD, ICASSP, ICML, and IJCAI.

• 09.2026 - Joined X-tech as a Machine Learning Quant Algorithm Engineer.
• 08.2026 - Three papers were accepted to EMNLP 2026: Decomposing Visual Histories with Vision-Language Agents (Main Conference), Demystifying Hidden-State Recurrence (Findings), and NUM2EVENT (Findings, co-first author).
• 07.2026 - When Efficiency Becomes Fragility: Exploiting Dynamic Routing Vulnerabilities in Adaptive UAV Tracking was accepted to the ACM MM 2026 Main Track.
• 06.2026 - First-author paper Dynamic-V2C was accepted to ECCV 2026.
• 06.2026 - Our HKUST(GZ) Deep Robotics Team / Nexus Robotics Lab won 1st place globally in WBCD 2026 Logistics Picking, held with ICRA 2026, and also placed first at the Shanghai site.
• 06.2026 - SANT-CBM was provisionally accepted to MICCAI 2026; I serve as the last and corresponding author. Our concept-consistent semi-supervised CBM paper was accepted to ICANN 2026; I serve as the corresponding author.
• 06.2026 - Free-T2M (Poster) and Delta Score Matters! (Oral) were accepted to the ACM MM 2026 BNI Track.
• 03.2026 - CAT was accepted to TMLR 2026. Project | Code
• 03.2026 - Towards Better Evaluation Metrics for Text-to-Motion Generation received the Best Paper Runner-Up Award at the ACM TheWebConf 2026 Workshop TIME.

More updates are available on my personal website.

• Decomposing Visual Histories with Vision-Language Agents: Hierarchical Temporal Guidance for Compositional Image Generation. EMNLP 2026 Main Conference, accepted. OpenReview
• Demystifying Hidden-State Recurrence: Switchable Latent Reasoning with On-Policy Reinforcement Learning. EMNLP 2026 Findings, accepted. OpenReview
• NUM2EVENT: Structured Event Inference from Numerical Time-Series. EMNLP 2026 Findings, accepted. Co-first author. OpenReview
• When Efficiency Becomes Fragility: Exploiting Dynamic Routing Vulnerabilities in Adaptive UAV Tracking. ACM MM 2026 Main Track, accepted.
• Dynamic-V2C: Editable and Continual Vision-to-Concept Bottleneck Models via Influence Functions. ECCV 2026, accepted. First author. Project
• SANT-CBM: Structurally-Aware and Noise-Tolerant Semi-supervised Concept Bottleneck Models. MICCAI 2026, provisionally accepted. Last and corresponding author. Project
• Concept-Consistent Semi-Supervised Learning for Concept Bottleneck Models via Confidence-Guided Pseudo-Label Propagation. ICANN 2026, accepted. Corresponding author. Paper
• Multimodal Deception in Explainable AI: Concept-Level Backdoor Attacks on Concept Bottleneck Models. TMLR 2026. Paper | Project | Code
• ACE: Attribution-Controlled Knowledge Editing for Multi-hop Factual Recall. ICLR 2026. Co-first author. Paper | Project | Code
• Towards Reliable Time Series Forecasting under Future Uncertainty: Ambiguity and Novelty Rejection Mechanisms. ICASSP 2026. Co-first author. Paper | Project
• Free-T2M: Frequency-Aware Coarse-to-Fine Text-to-Motion Generation. ACM MM 2026 BNI Track, Poster. OpenReview
• Delta Score Matters! Spatial Adaptive Multi Guidance in Diffusion Models. ACM MM 2026 BNI Track, Oral. arXiv

• Learning New Concepts, Remembering the Old: Continual Learning for Multimodal Concept Bottleneck Models. ACM MM 2025 BNI Track. First author. Paper | Project | Code
• From Guesswork to Guarantee: Towards Faithful Multimedia Web Forecasting with TimeSieve. ACM MM 2025. First author. Paper | Project | Code
• DRIVE: Dependable Robust Interpretable Visionary Ensemble Framework in Autonomous Driving. ICRA 2025. First author. Paper | Project | Code
• Stable Vision Concept Transformers for Medical Diagnosis. ECML-PKDD 2025. Co-first author. Paper | Project | Code
• PEPL: Precision-Enhanced Pseudo-Labeling for Fine-Grained Image Classification in Semi-Supervised Learning. ICASSP 2025. Co-first author. Paper | Project
• Faithful Vision-Language Interpretation via Concept Bottleneck Models. ICLR 2024. First author. Paper | Project | Code
• Towards Multi-dimensional Explanation Alignment for Medical Classification. NeurIPS 2024. Co-first author. Paper | Project | Code

See the complete publication list for all papers and co-authors.

I received more than 40 university-level awards spanning academic competitions, social practice, innovation, sports, volunteer service, and scholarships.

• Conference and Journal Reviewer: ECAI, ICML, KDD, ICLR, CVPR, ICCV, NeurIPS, ACM MM, IJCAI, Expert Systems with Applications, and others.
• Class Monitor, Chongxin College: led a class recognized as a Shandong Provincial Excellent Class and one of Shandong University's Top Ten Classes.
• Outstanding Volunteer: 130 hours of volunteer service.

Yutao Yue | Zhi Liu | Zhijiang Guo | Lijie Hu | Jiayu Yang | Shengen Wu | Jiaqi Tang | Wenshuo Chen | Mingqian Liao | Ninghui Feng | Yuxuan Fan

More collaborators and project details are listed on my personal website.