CV

Education

• Ph.D. in Computer Science, University of Sydney, 2025.03-2027.09 (expected)
  • Advisor: Prof. Chang Xu
• B.S. in Software Engineering, Beihang University, 2019.09-2023.06
  • Average score: 89/100

Work experience

• 2025.04-2025.10: Embodied LLM Algorithm Research Intern
  • AgiBot Embodied Research Center
  • Vision-Language-Action model status monitoring, error handling, and test-time scaling.
• 2023.02-2025.02: Full-time Algorithm Researcher
  • SenseTime Research
  • Multimodal LLM hallucination mitigation and lightweight diffusion-based image generation.

Skills

• Vision-Language-Action models, embodied AI, agents, multimodal LLMs, and diffusion models.
• Python, C, PyTorch, JAX, DeepSpeed, Hugging Face Transformers, and Diffusers.

Highlights

• 4 first-author papers accepted by CCF A / Core A* conferences, including 2 Oral papers.
• National Undergraduate Innovation and Entrepreneurship Training Program: 2 excellent completions as project leader.
• Beihang University Science and Technology Innovation Scholarship: first prize, 3 times.

Latest CV

• Download the latest CV PDF

Publications

VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Model

Li W, et al. VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Model. International Conference on Machine Learning (ICML), 2026. CCF A. First author.

Sentinel-VLA: A Metacognitive VLA Model with Active Status Monitoring for Dynamic Reasoning and Error Recovery

Li W, et al. Sentinel-VLA: A Metacognitive VLA Model with Active Status Monitoring for Dynamic Reasoning and Error Recovery. International Conference on Machine Learning (ICML), 2026. CCF A. First author.

Identify, Isolate, and Purge: Mitigating Hallucinations in LVLMs via Self-Evolving Distillation

Li W, et al. Identify, Isolate, and Purge: Mitigating Hallucinations in LVLMs via Self-Evolving Distillation. ACM International Conference on Multimedia (ACM MM), 2025. Oral, CCF A. First author.

Adaptive Training Meets Progressive Scaling: Elevating Efficiency in Diffusion Models

Li W, Su X, You S, et al. Adaptive Training Meets Progressive Scaling: Elevating Efficiency in Diffusion Models. IEEE International Conference on Multimedia and Expo (ICME), 2025.

DiffNAS: Bootstrapping Diffusion Models by Prompting for Better Architectures

Li W, Su X, You S, et al. DiffNAS: Bootstrapping Diffusion Models by Prompting for Better Architectures. IEEE International Conference on Data Mining (ICDM), 2023. Oral, Core A*.