Lichuan Xiang | 向力川

I am an Honorary Research Fellow at the University of Warwick and a Founding Research Scientist at Collov.ai. My research focuses on efficient Machine Learning systems, with expertise in neural architecture design, model compression, and automated ML systems that operate under resource constraints.

Research Interests

My work falls into these main areas:

1. Neural Architecture Search (NAS): I develop methods that can automatically discover optimal neural network architectures for specific tasks, with a focus on efficiency and performance.

2. Model Compression and Pruning: My research explores novel techniques to reduce the computational requirements of deep learning models while maintaining accuracy, particularly focusing on pruning at initialization.

3. Efficient Deep Learning: I work on creating algorithms that significantly reduce ML computation across the full lifecycle from architecture discovery to training and inference.

4. Controllable Generation with Diffusion Models: I develop methods for precise control of diffusion-based generative models, enabling spatially-aware object manipulation that follows photorealistic principles and physical constraints. My research focuses on computation-aware architectures for diffusion models that maintain quality while improving efficiency.

5. Agentic Systems: I research autonomous AI systems that can plan, reason, and execute complex tasks through dynamic tool use and environment interaction. This includes developing architectures that support self-directed learning and decision-making in multi-step reasoning tasks.

Current Work

At Collov.ai, I lead several key projects:

• Indoor 3D space reconstruction from single images
• Diffusion model training for image editing tasks
• Performance optimization of diffusion models
• Development of virtual staging tools for interior design

At the University of Warwick, I continue my research in efficient ML systems and supervise doctoral students working on related topics.

Recent Achievements

• FlexControl Accepted in ICML2025
• 🏆 Best Paper Award at AutoML Conference 2023 for our work on “Exploiting network compressibility and topology in zero-cost NAS”
• Published papers in top-tier conferences including ICML, ICLR, AAAI, and NeurIPS
• Successfully completed my Ph.D. in Computer Science from the University of Warwick (2024)

Please explore my CV, Publications, and Research Lab pages to learn more about my work. Feel free to contact me if you’re interested in collaborating.