Author

Ziyao Lin

Observed name: Lin, Ziyao

Institutions: University of Macau, South China University of Technology

Observed institutions: University of Macau, South China University of Technology

OpenAlex ID: A5127243448

Self-Cleaning and Captured Anyway: One Measured Primitive for Error in a Store an Agent Writes to Itself, and What a Falling Score Actually Measures

2026 · Chen, Wenhui, Chen, Jianlin, et al. · arXiv (Cornell University)

The Architect, the Adversary, and the Judge: Closed-Loop Generation of Standards-Aligned Assessment Items at Scale

2026 · Chen, Wenhui, Lin, Ziyao, et al. · arXiv (Cornell University)

A Judge Should Know What Changed:Construct Validity for LLM-as-a-Judge Evaluation

2026 · Chen, Jianlin, Chen, Wenhui, et al. · arXiv (Cornell University)

Budget-Constrained Embodied Perception: Four Resource Walls and a Pre-Registered Evaluation of Access-Structured Perception on Open Models at less than 31B

2026 · Lin, Defu, Chen, Wenhui, et al. · arXiv (Cornell University)

The Free-Recipe Limit: Every Recipe Effect Measures Which Premise of an Idealised Learner Broke

2026 · Chen, Wenhui, Chen, Jianlin, et al. · arXiv (Cornell University)

Auditing Discovery Claims: A Two-Sided Criterion for Agentic Science, with the Negative Side Decidable

2026 · Chen, Wenhui, Chen, Jianlin, et al. · arXiv (Cornell University)

Judging Is Not Enumerating: Silent Omissions in LLM-Authored Acceptable Sets

2026 · Chen, Wenhui, Chen, Jianlin, et al. · arXiv (Cornell University)

The Capability Convergence Hypothesis: Capability from Access Structure, Not Scale

2026 · Chen, Wenhui, Chen, Jianlin, et al. · arXiv (Cornell University)

DrivePTS: A Progressive Learning Framework with Textual and Structural Enhancement for Driving Scene Generation

2026 · Wang, Zhechao, Zeng, Yiming, et al. · arXiv (Cornell University)

Works count: 9

Citation count:

H-index:

I10-index: