Frontis-MA1: Tsinghua's AI4AI Model Aims for Recursive Self-Improvement in MLE
alex_verem · x · 2026-08-04
A Tsinghua University research team introduced Frontis-MA1, a 35B parameter model designed to explore Recursive Self-Improvement (RSI) in Machine Learning Engineering (MLE) through the "AI4AI" paradigm.
The researchers built OpenMLE, an open-source full-stack system featuring task environments, operator learning, and long-horizon search. They aligned post-training and inference around four atomic program-evolution operators (Draft, Improve, Debug, Crossover), trained via execution-grounded SFT and RL, coupling learning and evolution in a single loop.
Experiments show that on a single RTX 4090 (capped at 12GB VRAM) with a 12-hour budget, Frontis-MA1 improved the Medal Average on the MLE-Bench Lite benchmark from 39.39% to 60.61%, reaching 71.21% when combined with asynchronous experience priors.
Related event: Tsinghua Releases Frontis-MA1 for Recursive Self-Improvement(2 posts)→
More from coding & agent
- Quickly Turn Any Website into an API or MCP Server Using DevTools — dsp_ · 2026-08-04
- Anthropic Engineers Demo Multi-Agent Loops: Full App Built from Scratch in 40 Mins — mathemagie · 2026-08-04
- Running Agents in Prod for a Year: Auditability Trumps Model Capability — KimLikeJ · 2026-08-04
- Open-Source 'The Librarian' MCP Server Manages 3,000 Skills, Saves Tokens and Self-Optimizes — Open-Appeal-9747 · 2026-08-04
- DSPy Launches Experimental Flex Feature for Automated Code and Control Flow Optimization — lateinteraction · 2026-08-04
- OpenAI Hints Next-Gen Models Need More Compute, Codex May Shift to Cloud Agents — haider1 · 2026-08-04