ZYZhiyu Yang

Zhiyu Yang

Researcher-builder · AI agents, coding, and human–AI experience

kelvin.yangzhiyu@outlook.com github.com/KevinCL16 Google Scholar
Print / Save PDF Back to homepage

Education

University of Texas at Dallas

PhD in Computer Science

Beijing Language and Culture University

M.Eng. in Computer Science and Technology

Sichuan University

B.Eng. in Computer Science and Technology

Experience

Research Assistant · Singapore Management University

Worked with Yang Deng on LLM benchmarks for multi-hop and multi-bug data-science code debugging.

Research Intern · THUNLP, Tsinghua University

Worked on MatPlotAgent for agentic scientific visualization and UltraLink for multilingual supervised fine-tuning data.

Research Intern · ModelBest & OpenBMB

Worked on LLM×MapReduce for adapting regular LLMs to long-context inputs.

Selected publications

  1. Yang, Zhiyu, et al. “Why Stop at One Error? Benchmarking LLMs as Data Science Code Debuggers for Multi-Hop and Multi-Bug Errors.” EMNLP 2025 Main — Oral.
  2. Yang, Zhiyu, et al. “MatPlotAgent: Method and Evaluation for LLM-Based Agentic Scientific Data Visualization.” ACL 2024 Findings.
  3. Yang, Hao, Yang, Zhiyu, et al. “CompactRAG: Reducing LLM Calls and Token Overhead in Multi-Hop Question Answering.” WWW 2026.
  4. Yang, Zhiyu, et al. “Enhancing Free-Form Table Question Answering Models by Distilling Relevant-Cell-Based Rationales.” CCL 2024.
  5. Wang, Haoyu, Yang, Zhiyu, et al. “UltraLink: An Open-Source Knowledge-Enhanced Multilingual Supervised Fine-tuning Dataset.” ACL 2024.