Yiming Zheng

About Me

I am now diving in Reinforcement Learning/Reasoning Method/Models. Training myself through real-world interactions. I am also very pleased to be contacted for academic collaboration. It is always a great opportunity to engage in discussions and work together on research that can contribute to the advancement of knowledge. I look forward to exploring potential projects and sharing insights with like-minded scholars.

Research Interests

• Large Language Models:
  • Post-training Techniques: Reinforcement Learning with Human Feedback (RLHF), focusing on advanced policy optimization algorithms.
  • Reasoning and Planning: Enhancing model inference capabilities, including multi-step reasoning, tool-use, and decision-making in dynamic environments.
  • Tool-Augmented LLMs: Integration with external tools and APIs to improve task completion and real-world interaction abilities.
  • Embodied Agents for Gaming: Developing LLM-driven agents capable of understanding, acting, and adapting within interactive gaming environments.
• Multimodal Generation:
  • Controllable Text-to-Speech (TTS): Fine-grained control over speech synthesis, emphasizing semantic alignment, prosodic variation, and expressive rendering from textual input.

News

• [May. 2025] I have commenced a research intern position as an Application Researcher at Turing Lab, Tencent IEG, focusing on applied research in focusing on AI for games, text-to-speech (TTS), and related applications, under the supervision of Prof. Junle Wang.
• [Oct. 2024] I am now joining ZHIPU AI as an NLP Algorithm Intern in the LingXin Department, supervised by Prof. Minlie Huang.
• [Oct. 2024] I have received an offer of exemption from examinations from SIAT.
• [Aug. 2024] Remote RA at HKBU TMLR Group.

Publications

Services

Powered by Jekyll and Minimal Light theme.