I'm a Junior student at SJTU, majoring in Artificial Intelligence. I have a strong interest in Machine Learning, Deep Learning and especially RLVR of LLM. Now I'm a research intern at DAIR, focusing on improving the efficiency and performance of (multimodal) LLMs.
A novel RLVR algorithm that adaptively adjusts the weights of samples based on their difficulty levels, enhancing both learning efficiency and final performance.
An open-source reference in self-learning of artificial intelligence.
2025-PRESENT © Kinnari ✨