Publications

You can also find my articles on my Google Scholar profile.

Journal Articles

Decoding cortical folding patterns in marmosets using machine learning and large language model

Published in NeuroImage, 2025

Identification of genes with transcriptomic differences between concave and convex cortical patterns using machine learning and LLM.

Recommended citation: Yue Wu, Xuesong Gao, Zhengliang Liu, Pengcheng Wang, Zihao Wu, Yiwei Li, Tuo Zhang, Tianming Liu, Tao Liu, Xiao Li, Decoding cortical folding patterns in marmosets using machine learning and large language model, NeuroImage, Volume 308, 2025
Download Paper

Conference Papers

Entropy Regularized Process Reward Model

Published in TMLR, 2025

This paper proposes an Entropy-Regularized Process Reward Model (ER-PRM) to improve mathematical reasoning in large language models. The key novelty is formulating multi-step reasoning under an entropy-regularized Markov Decision Process framework, which balances reward optimization with preventing the policy from deviating too far from its initial distribution. The method derives process reward scores using a novel aggregation approach based on KL-regularized optimization, where rewards are computed as the logarithm of expected exponentiated rewards from completion trajectories sampled by the initial policy. This approach offers theoretical advantages including dual formulation flexibility (soft-max when sampling from initial policy, soft-min from optimal policy) and independence from the optimal policy during reward computation.

Recommended citation: Hanning Zhang*, Pengcheng Wang*, Shizhe Diao, Yong Lin, Rui Pan, Hanze Dong, Dylan Zhang, Pavlo Molchanov, & Tong Zhang (2025). Entropy-Regularized Process Reward Model. Transactions on Machine Learning Research.
Download Paper

Active Prompting with Chain-of-Thought for Large Language Models

Published in ACL, 2024

This paper proposes a new method, Active-Prompt, to adapt LLMs to different tasks with task-specific example prompts (annotated with human-designed CoT reasoning).

Recommended citation: Shizhe Diao, Pengcheng Wang, Yong Lin, Rui Pan, Xiang Liu, and Tong Zhang. 2024. Active Prompting with Chain-of-Thought for Large Language Models. In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1330–1350, Bangkok, Thailand. Association for Computational Linguistics.
Download Paper

Preprints

AgentSPEX: An Agent SPecification and EXecution Language

Published in arXiv preprint, 2026

AgentSPEX is an Agent SPecification and EXecution Language for specifying LLM-agent workflows with explicit control flow and modular structure, along with a customizable agent harness. It supports typed steps, branching and loops, parallel execution, reusable submodules, and explicit state management; workflows execute within an agent harness that provides tool access, a sandboxed virtual environment, and support for checkpointing, verification, and logging. We also provide a visual editor with synchronized graph and workflow views for authoring and inspection, ship ready-to-use agents for deep research and scientific research, evaluate AgentSPEX on 7 benchmarks, and show through a user study that AgentSPEX offers a more interpretable and accessible workflow-authoring paradigm than a popular existing agent framework. Project page: agentspex.ai. Code: ScaleML/AgentSPEX.

Recommended citation: Pengcheng Wang*, Jerry Huang*, Jiarui Yao*, Rui Pan, Peizhi Niu, Yaowenqi Liu, Ruida Wang, Renhao Lu, Yuwei Guo, & Tong Zhang (2026). AgentSPEX: An Agent SPecification and EXecution Language. arXiv preprint arXiv:2604.13346.
Download Paper