Yixin Ye

I’m a third year undergraduate student from Shanghai Jiao Tong University, advised by Prof.Pengfei Liu at GAlR Lab. Previously, I was an undergraduate researcher advised by Prof.Dequan Wang.
My research interests mainly lie in natural language processing, complex reasoning, and fundamental research of LLMs. I am now focusing on enhancing LLMs’ reasoning capabilities through innovative approaches.
I have experience in developing and training large language models, with a focus on pretraining and fine-tuning techniques. My background also includes work on multimodal learning systems and addressing alignment challenges in multimodal AI architectures.
news
Feb 13, 2025 | We are honored to have been invited to present a talk about LIMO at Synced (机器之心). |
---|---|
Feb 10, 2025 | We released a real-time benchmark for math reasoning models on AIME 2025! |
selected publications
- arxiv
- arxiv
- ECCVDissecting Dissonance: Benchmarking Large Multimodal Models Against Self-Contradictory InstructionsIn Computer Vision – ECCV 2024, 2025
- ECCVLost in Translation: Latent Concept Misalignment in Text-to-Image Diffusion ModelsIn Computer Vision – ECCV 2024, 2025
misc
In my free time, I enjoy indie rock music 🎸, traditional strength training 🏋️ and amateur long-distance running 🏃♀️.