Hi 👋, I'm Jingxuan! I'm a PhD student in the Department of Electrical and Electronic Engineering at Imperial College London, working on unified multimodal models. My research explores how large vision-language models can be adapted to data-scarce domains such as clinical imaging, and how autonomous LLM agents can take over more of the scientific research loop. I have published at top venues including CVPR, ICCV, ICML, and NeurIPS, and serve as a reviewer for ICML, AAAI, CVPR, ICLR, NeurIPS, ECCV, and MICCAI.

🏠 jingxuan.uk · 🎓 Google Scholar · 🐦 X · ✉️ j.kang26@ic.ac.uk · 💬 WeChat: Romani0618

• AI 科研不完全指北：Top10 PhD 烧了上万刀的真实经验。Context Is All You Need
• AI 科研不完全指北第二期：如何 Propose 一个 idea
• 我的 AI 订阅清单：一份主观推荐
• REPA：让 DiT 训得快一点 (paper note)

More posts →

• 🚩 2026.06 — Selected as a Gold Reviewer for ICML 2026
• 🚩 2026.04 — Started my PhD at Imperial College London
• 2026.03 — Paper on saliency-guided prompt distillation for SAM accepted to CVPR 2026 (Findings); co-authored work on infrared video super-resolution also accepted to CVPR 2026
• 2025.05 — Paper on attention-based selection for vision-language models accepted to ICML 2025

• Learning from Noisy Prompts: Saliency-Guided Prompt Distillation for Robust Segmentation with SAM — CVPR 2026, Findings (paper)
• From Local Details to Global Context: Advancing Vision-Language Models with Attention-Based Selection — ICML 2025 (paper)
• Translating Simulation Images to X-ray Images via Multi-Scale Semantic Matching — MICCAI Workshop 2024 (paper)
• Style Transfer Meets Super-Resolution: Advancing Unpaired Infrared-to-Visible Image Translation with Detail Enhancement — ACM MM 2023 (paper)

Full list: jingxuan.uk/#publications