# 混元多模态强化学习（RL）算法研究员（北京/上海）

External opening: you cannot apply through AI Native Match; apply on the official site.

Apply through the official channel. The full external receipt stays in your workspace. When signed in, three minimal event types explicitly recorded through the CLI are reported by default: selection, self-reported application and channel observation. Browsing or saving a role does not report an event; external applications do not appear under My applications.

- Company: 腾讯 (Stage unknown)
- Locations: Unknown
- Role: Algorithm
- Apply on the official site: https://careers.tencent.com/jobdesc.html?postId=2034532284522196992

## Role description from source

1.针对多模态模型开展强化学习算法研究，包括面向图像、视频生成的扩散模型，面向多模态理解的自回归模型，以及前沿统一多模态框架；
2.设计并开发强化学习训练框架与奖励建模策略，实现高效的大规模训练，提升训练稳定性，并解决奖励作弊等相关问题；
3.探索下一代强化学习范式，使其能更直接、更高效地从环境反馈中学习。
职位要求
1.本科及以上学历，计算机科学或相关专业；
2.具备出色的科研能力，在ICML、NeurIPS、ICLR、CVPR、ICCV、ECCV、SIGGRAPH等顶级会议发表过论文；
3.扎实的工程与编程能力，具备深度学习系统实现、模型训练与推理优化、CPU/GPU 加速、分布式训练与推理经验；
4.有扩散模型、自回归模型、文生图 / 文生视频生成相关经验者优先；
5.有 ACM/NOIP（信息学竞赛）参赛经历者优先。
工作年限（官方顶部字段）
两年以上工作经验

---
opening_id: 445dd5aa-e57f-4026-8789-5f5fe91c0820

`anm openings save --id 445dd5aa-e57f-4026-8789-5f5fe91c0820`