PhD Candidate at Carnegie Mellon University

Danqing Wang

Strategic planning and reasoning for LLM agents

About

I am a PhD candidate at the Language Technologies Institute at Carnegie Mellon University, advised by Prof. Lei Li. My research develops strategic planning and reasoning methods for LLM agents, with an emphasis on multi-agent collaboration, competition, communication, and safety. I also work on AI for science, particularly drug discovery.

Before joining CMU, I was a PhD student and member of the UCSB NLP Group at UC Santa Barbara, and a full-time researcher at ByteDance AI Lab. I received my master’s and bachelor’s degrees in computer science from Fudan University, where I was advised by Prof. Xipeng Qiu and Prof. Xuanjing Huang.

I am on the 2026-27 job market for full-time positions starting in 2027. Please get in touch if my research may be a fit for your team.

Selected Agent Research

EMNLP 2023Collaboration

Learning from Mistakes via Cooperative Study Assistant for Large Language Models

A cooperative study assistant records prior mistakes and retrieves tailored guidance to prevent recurring reasoning errors.

Project Paper Code

ICLR 2026Competition

Strategic Planning and Rationalizing on Trees Make LLMs Better Debaters

Rehearsal and debate-flow trees help LLMs anticipate attacks, track arguments, and allocate speaking time strategically.

Project Paper Code

CoLM 2026Communication

Learning to Interrupt in Language-based Multi-agent Communication

HANDRAISER learns when listeners should interrupt, reducing communication cost by 32.2% while preserving or improving task performance.

Paper

ICML 2026Safety

Is Vibe Coding Safe? Benchmarking Vulnerability of Agent-Generated Code in Real-World Tasks

A benchmark of 187 real-world feature requests exposes the gap between functional and secure code produced by coding agents.

Project Paper Code

View all publications → Explore projects →

Recognition

2025 Amazon AI PhD Fellowship

Selected Experience

May 2026-present: Research Scientist Intern, Microsoft Research
- Mentors: Baolin Peng and Jianfeng Gao
May-Oct 2025: Research Scientist Intern, Meta AI (FAIR), working on multi-agent communication
- Mentors: Ansong Ni and Asli Celikyilmaz
Mar-May 2024: Research Scientist Intern, Alibaba (Qwen), working on LLM reasoning
- Mentor: Jianxin Ma
Jun 2023-Jan 2024: Research Scientist Intern, Meta AI (FAIR), working on personalized alignment
- Mentor: Yuandong Tian
Apr 2020-Aug 2022: Researcher, ByteDance AI Lab, working on text generation and AI for drug design
- Mentors: Jiaze Chen, Hao Zhou, and Lei Li

Education

Jan 2024-present: PhD, Language Technologies Institute, Carnegie Mellon University (expected June 2027)
Sep 2022-Dec 2023: PhD studies in Computer Science, UC Santa Barbara (transferred to CMU)
2018-2021: MS in Computer Science, Fudan University
2014-2018: BS in Computer Science and Technology, Fudan University

Selected Service & Teaching

Program committees: ICML, NeurIPS, ICLR, ACL, NAACL, EMNLP, AAAI, KDD, and NLPCC
Organizer: Southern California NLP Symposium 2022
Teaching: LLM Systems at CMU; Deep Learning and Machine Learning at UCSB

More details are available in my full CV.