Hao Wang 王昊

B.Eng. in Computer Science, University of Science and Technology of China

Incoming M.S. Student, MIRA Lab

I am a final-year undergraduate student in Computer Science at the University of Science and Technology of China (USTC), and I will join MIRA Lab as a graduate student in 2027.

My research focuses on LLM post-training, especially agentic reinforcement learning — training language agents to plan, use tools, and recover from failures. I am also interested in reliable evaluation of agent behavior, and in building agents that are genuinely useful outside a demo.

Illustrated portrait of Hao Wang

News

Earlier news
  • Awarded Gold Medal in the 2025 iGEM competition! Explore our project wiki.
  • Secured Second Prize at RoboGame! Check out our Github team.

Education & Experience

Education

Experience

Selected Publications

All publications
Figure from ReLiB-100k: a real-world 100k-scale dataset and comprehensive benchmark for capacity estimation of retired lithium-ion batteries

ReLiB-100k: a real-world 100k-scale dataset and comprehensive benchmark for capacity estimation of retired lithium-ion batteries

Yunlong Liu, Pengfei Zhou, Hao Wang, Zhiqiang Qin, Ji Wu, Wangqiu Zhou, Gang Luo, Qi Song, Xiangyang Li

Data Mining and Knowledge Discovery (DMKD) 2026

Code Paper

Selected Projects

All projects

dsh-llm-pool

A DSH plugin for building configurable model pools from multiple providers — each pool acts as one virtual provider, with session stickiness, load balancing, and pre-first-token failover.

Project page Code

dsh-auto-workflow

A DeepSeek Harness plugin for persistent, concurrent workflows — agents, commands, HTTP calls, and Human decisions in one durable node graph.

Project page Code

Mirally

A personal agent designed to go beyond chat — with long-term context, multi-step planning, and tool use.

Project page

Nine Sword Techniques

A Slay the Spire mod featuring nine sword techniques, the Muzixi character, and new game content.

Project page Code Steam Workshop

Selected Honors

All honors