Shangbin Feng

CVEmailScholarGitHubX

I'm on the 2026-2027 faculty job market.

I am a final-year PhD candidate at the University of Washington, working with Yulia Tsvetkov. Previously, I was a research intern at NVIDIA and Google.

Photo of Shangbin Feng

Compositional Intelligence for Participatory AI

Participatory AI: contributors each bring their own model, and the models are composed into one modular system, across the Train, Align, Adapt, and Deploy stages.

Today's AI is built in two ways: closed AI, developed by a few companies that decide what data to train on and whose values to encode, and open AI, open mostly in outcome, releasing weights and sometimes data and code, while decisions over what goes into the model still rest with one team. Both have driven remarkable progress, yet in both, the people AI serves have little say in how it is built.

My research charts a third paradigm: participatory AI, modular systems built bottom-up from the contributions of diverse stakeholders, who keep ownership of what they contribute and agency over how it is used. Its foundation is compositional intelligence, where independently built modular components collaborate so that the whole achieves capabilities beyond what any part was built for. Participatory AI raises new technical questions across the model lifecycle:

I am also broadly interested in knowledge and factuality, networks and structures, social NLP, multi-agent systems, AI safety, and applications in scientific domains.

Honors and awards

Selected publications

* equal contribution  ·  full list on Google Scholar

  1. Scaling Participation in Modular AI Systems
    Shangbin Feng, Yike Wang, Weijia Shi, Luke Zettlemoyer, Yejin Choi, Yulia Tsvetkov
    Under review at a journal, 2026 papercode
  2. Compositional RL: Evaluating and Training LLMs for Collaborative Reasoning
    Shangbin Feng, Yulia Tsvetkov, Jan Kautz, David Acuna
    Under review, 2026
  3. Multi-LLM Collaborative Alignment via Stackelberg Games
    Christina Hahn*, Shangbin Feng*, Dean Light, Swastik Roy, Hila Gonen, Yulia Tsvetkov
    Under review, 2026 papercode
  4. MoCo: A One-Stop Shop for Model Collaboration Research
    Shangbin Feng*, Yuyang Bai*, Ziyuan Yang*, Yike Wang, Zhaoxuan Tan, Jiajie Yan, Zhenyu Lei, Wenxuan Ding, Weijia Shi, Haojin Wang, Zhenting Qi, Yuru Jiang, Heng Wang, Chengsong Huang, Yu Fei, Jihan Yao, Yilun Du, Luke Zettlemoyer, Yejin Choi, Yulia Tsvetkov
    Under review, 2026 papercode
  5. Among Us: Measuring and Mitigating Malicious Contributions in Model Collaboration Systems
    Ziyuan Yang*, Wenxuan Ding*, Shangbin Feng*, Yulia Tsvetkov
    ACL 2026, Senior Area Chair Award, Oral papercode
  6. When One LLM Drools, Multi-LLM Collaboration Rules
    Shangbin Feng, Wenxuan Ding, Alisa Liu, Zifeng Wang, Weijia Shi, Yike Wang, Zejiang Shen, Xiaochuang Han, Hunter Lang, Chen-Yu Lee, Tomas Pfister, Yejin Choi, Yulia Tsvetkov
    ACL 2026, Oral paper
  7. Sparta Alignment: Collectively Aligning Multiple Language Models through Combat
    Yuru Jiang*, Wenxuan Ding*, Shangbin Feng*, Greg Durrett, Yulia Tsvetkov
    NeurIPS 2025 papercode
  8. Heterogeneous Swarms: Jointly Optimizing Model Roles and Weights for Multi-LLM Systems
    Shangbin Feng, Zifeng Wang, Palash Goyal, Yike Wang, Weijia Shi, Huang Xia, Hamid Palangi, Luke Zettlemoyer, Yulia Tsvetkov, Chen-Yu Lee, Tomas Pfister
    NeurIPS 2025 · Integrated into Google's Gemini Enterprise Agent Designer papercode
  9. Model Swarms: Collaborative Search to Adapt LLM Experts via Swarm Intelligence
    Shangbin Feng, Zifeng Wang, Yike Wang, Sayna Ebrahimi, Hamid Palangi, Lesly Miculicich, Achin Kulshrestha, Nathalie Rauschmayr, Yejin Choi, Yulia Tsvetkov, Chen-Yu Lee, Tomas Pfister
    ICML 2025 papercode
  10. Modular Pluralism: Pluralistic Alignment via Multi-LLM Collaboration
    Shangbin Feng, Taylor Sorensen, Yuhan Liu, Jillian Fisher, Chan Young Park, Yejin Choi, Yulia Tsvetkov
    EMNLP 2024 papercode
  11. Don't Hallucinate, Abstain: Identifying LLM Knowledge Gaps via Multi-LLM Collaboration
    Shangbin Feng, Weijia Shi, Yike Wang, Wenxuan Ding, Vidhisha Balachandran, Yulia Tsvetkov
    ACL 2024, Outstanding Paper Award, Senior Area Chair Award papercode
  12. Knowledge Card: Filling LLMs' Knowledge Gaps with Plug-in Specialized Language Models
    Shangbin Feng, Weijia Shi, Yuyang Bai, Vidhisha Balachandran, Tianxing He, Yulia Tsvetkov
    ICLR 2024, Oral papercode
  13. Can Language Models Solve Graph Problems in Natural Language?
    Heng Wang*, Shangbin Feng*, Tianxing He, Zhaoxuan Tan, Xiaochuang Han, Yulia Tsvetkov
    NeurIPS 2023, Spotlight papercode
  14. From Pretraining Data to Language Models to Downstream Tasks: Tracking the Trails of Political Biases Leading to Unfair NLP Models
    Shangbin Feng, Chan Young Park, Yuhan Liu, Yulia Tsvetkov
    ACL 2023, Best Paper Award · Covered by The Washington Post and MIT Technology Review papercode

Mentoring

I have been fortunate to work with and mentor these brilliant students: