Bo Peng (彭博)
I teach AI to ask why, not just what.
I’m a third-year PhD student at Shanghai Jiao Tong University, working at the intersection of causal reasoning and large language models. I am co-supervised by Prof. Yu Qiao and Dr. Chaochao Lu, with affiliations at both SJTU and Shanghai AI Lab.
My research pursues a single thread: equipping AI systems with the ability to reason about cause and effect, not just pattern-match on correlations. This spans:
- Causal Evaluation — benchmarking whether LLMs/VLMs truly understand causal relationships (CELLO, CaLM)
- Causal Discovery — scaling neural methods to discover causal structures from data (CauScale, CauScientist)
- Personalization & Generation — building models that adapt to users and generate conditioned on causal structure (IP-Dialog, ConditionVideo)
Feel free to reach out if you’d like to collaborate or just chat about causality and AI.
News
- [May 2026] CauScale accepted to ICML 2026!
- [Sep 2025] IP-Dialog accepted to EMNLP 2025 Findings.
- [Sep 2024] CELLO accepted to EMNLP 2024.
- [Dec 2023] ConditionVideo accepted to AAAI 2024.
Publications
CauScale: Neural Causal Discovery at Scale
Bo Peng, Sirui Chen, Jian Tian, Yu Qiao, Chaochao Lu
ICML 2026 · Paper · Code
CauScientist: Teaching LLMs to Respect Data for Causal Discovery
Bo Peng, Sirui Chen, Libo Xu, Chaochao Lu
arXiv preprint, 2026 · Paper · Code
IP-Dialog: Evaluating Implicit Personalization in Dialogue Systems with Synthetic Data
Bo Peng*, Zhiheng Wang*, Heyang Gong, Chaochao Lu
EMNLP Findings, 2025 · Paper · ACL Anthology
CauSight: Learning to Supersense for Visual Causal Discovery
Yan Zhang, Meiqi Chen, Sirui Chen, Bo Peng, Yan Zhang, Tao Li, Chaochao Lu
arXiv preprint, 2025 · Paper
Causal Evaluation of Language Models (CaLM)
Sirui Chen*, Bo Peng*, Meiqi Chen, Ruiqi Wang, Mengying Xu, Xingyu Zeng, Rui Zhao, Shengjie Zhao, Yu Qiao, Chaochao Lu
arXiv preprint, 2024 · Project Page
CELLO: Causal Evaluation of Large Vision-Language Models
Meiqi Chen, Bo Peng, Yan Zhang, Chaochao Lu
EMNLP, 2024 · Paper · 57 citations
ConditionVideo: Training-Free Condition-Guided Text-to-Video Generation
Bo Peng, Xinyuan Chen*, Yaohui Wang, Chaochao Lu, Yu Qiao
AAAI, 2024 · Project Page · 24 citations
Education
- Ph.D., Shanghai Jiao Tong University, 2023 – 2028 (expected)
- B.S., Shanghai Jiao Tong University, 2019 – 2023