Linghao Zhang

Technical Staff @ Xiaomi MiMo

About

I'm doing post-training at Xiaomi MiMo. Previously, I worked as a research intern at Microsoft Research, advised by Shilin He and Yu Kang.

I build something verifiable and scalable for code. I also work on coding agents & RL.

News

1 Paper accepted to NeurIPS 2025.
SWE-bench-Live - Tracking real coding capabilities in LLM and agents.
DI-Bench accepted to ACL Findings 2025.

Research

MiMo-V2-Flash

A powerful, efficient, and ultra-fast foundation language model.

LLM-Core Xiaomi
Report Blog HuggingFace

SWE-bench Goes Live!

Scaling data for SWE tasks by automate the setup of execution environments.

Linghao Zhang, Shilin He, Chaoyun Zhang, Yu Kang, Bowen Li, Chengxing Xie, Junhao Wang, Maoquan Wang, Yufan Huang, Shengyu Fu, Elsie Nallipogu, Qingwei Lin, Yingnong Dang, Saravan Rajmohan, Dongmei Zhang
NeurIPS 2025
Paper Code Project

DI-Bench: Benchmarking Large Language Models on Dependency Inference with Testable Repositories

A CI-based benchmark curation pipeline for software engineering tasks at the repository level.

Linghao Zhang, Junhao Wang, Shilin He, Chaoyun Zhang, Yu Kang, Bowen Li, Jiaheng Wen, Chengxing Xie, Maoquan Wang, Yufan Huang, Elsie Nallipogu, Qingwei Lin, Yingnong Dang, Saravan Rajmohan, Dongmei Zhang, Qi Zhang
ACL Findings 2025
Paper Code