Xinyu Huang (黄新宇)

I am a final-year Ph.D. student at the School of Computer Science, Fudan University, advised by Prof. Rui Feng and Prof. Yuejie Zhang. Currently, I'm visiting the MMLab at Nanyang Technological University under the supervision of Prof. Ziwei Liu from March to September 2025.

From 2020 to 2024, I researched on visual perception and created the Recognize Anything Model (RAM) Family: a series of open-source and powerful image perception models that exceed CLIP by more than 20 points in fine-grained perception. This work was done at OPPO Research Institute & IDEA, where I was fortunate to collaborate with Youcai Zhang, Prof. Yandong Guo, and Prof. Lei Zhang.

Since 2024, I focus on the the full process construction of large multimodal models, such as GPT-4o, working with a talented team at TikTok.

I expect to graduate in September 2025. I am open to both academic and industry research positions. Please feel free to download my Resume and do not hesitate to email me if you're interested :)

Email  /  Scholar  /  Github  /  Zhihu

RAM++ is the next generation of RAM, which can recognize any category with high accuracy, including both predefined common categories and diverse open-set categories.

RAM is an image tagging model, which can recognize any common category with high accuracy.

Tag2Text is a vision-language model guided by tagging, which can support tagging and comprehensive captioning simultaneously.

We propose IDEA to provide more explicit textual supervision (including multiple valuable tags and texts composed by multiple tags) for visual models.

Multi-label learning in the presence of missing labels(MLML) is a challenging problem. We propose two simple yet effective methods via robust loss design based on an observation.

We provide Recognize Anything Model Family (RAM) demonstrating superior image recognition ability!

RAM Faimly marry Grounded-SAM, which can automatically recognize, detect, and segment for an image! RAM Family showcases powerful image recognition capabilities!