QwQ-32B: Embracing the Power of Reinforcement Learning

Canonical link

https://arxiv.org/abs/2503.20735

Connected researchers

Junyang Lin (Justin Lin) is a researcher and open-source maintainer known for the Qwen family of models. His public profiles list interests in LLMs, AI agents, multimodal learning, long-horizon reasoning, world models, and reinforcement learning; multiple March 2026 news reports said he stepped down from the Qwen tech lead role.

Alibaba senior technology leader and researcher associated with Qwen. Public profiles list him with Alibaba Group, and official Alibaba Cloud coverage identifies him as a chief technology officer leading large-model work.

Researcher at Alibaba Group working on natural language processing and multimodal AI.

Research scientist in Tongyi Lab whose public profile highlights work on speech processing, machine learning, and multimodal large language models.

PhD candidate at Shanghai Jiao Tong University focused on reinforcement learning, large language models, and coding agents; coauthor of QwQ-32B.

Canonical link

Junyang Lin

Jingren Zhou

Fei Huang

Yongqi Wang

Yuxiang Zheng