Dongdong Hua

Dongdong Hua

I am a PhD Candidate in Computer Science at Zhejiang University advised by Yang Yang. I study continuously self-improving agents for long-horizon tasks, using open-ended environments to evaluate their capacity for sustained learning and adaptation at scale, with the goal of enabling real-world applications.

Email Google Scholar GitHub
Publications
System overview diagram for PTCG-Bench
EMNLP Findings,

PTCG-Bench: Can LLM Agents Master Pokémon Trading Card Game?

Dongdong Hua, Yifei Sun, Renhong Huang, Feng Gao, Chunping Wang, Yang Yang

PTCG-Bench evaluates LLM agents in Pokémon Trading Card Game environments, measuring both single-environment decision-making and self-evolution through accumulated experience.

arXiv PDF Project
System overview diagram for RAT
EMNLP,

RAT: RunAnyThing via Fully Automated Environment Configuration

Renhong Huang, Dongdong Hua, Yifei Sun, Sitao Ding, Hanyang Yuan, Daixin Wang, Yang Yang

RAT proposes a language-agnostic framework for automated environment configuration on arbitrary repositories and introduces RATBench, a benchmark for evaluating environment setup across heterogeneous real-world repositories.

arXiv PDF Project