arXiv 2609.22068 · cs.AI · 2026-09-18
CodeMidas: Scaling Agentic Coding RL Environments from Code Itself
Bowen Ye, Lei Li, Shicheng Li, Zihao Yue, Linghao Zhang, Hanglong Lv, Yuanxin Liu, Wenhan Ma and 11 more
Xiaomi MiMo
Archived · 2026-09 2026-W39 #8 Added on 2026-10-21
One-line takeaway
CodeMidas turns implemented functionality in codebases into executable coding RL environments using only source code.
CodeMidas 只以源代码为输入，自动把开源代码库中的功能转成可执行的编程 RL 环境。
Read on arXiv PDF Hugging Face discussion Project page
Topics & disciplines
More on Code & SWE
- AesCode: Aesthetic Code Generation with Decoupled Cross-Modal Rewards ▲137
- LEGO-Anything: Coding Agents for 3D Scene Reconstruction ▲129
- 3DCodeVerse: Exploring 3D Worlds as Code ▲117
- Groupwise Agentic Grading and Advantage Redistribution for Code Agent RL ▲95
- Last Secure Code Benchmark: Is the Code You Vibe Secure? ▲76
About this data
Metadata from the arXiv API (verified); upvotes from Hugging Face Daily Papers; citations from Semantic Scholar. Last refreshed 2026-10-21.
The takeaway is written by AIMap from the abstract; the abstract itself is not reproduced — read the original.