frontier models
A Machine-Learning Agent Learns Between Experiments
MLExpert assigns shared rewards to the decisions between costly experiment results instead of waiting for an entire engineering run to end.
Summary
MLExpert assigns shared rewards to the decisions between costly experiment results instead of waiting for an entire engineering run to end.
Long machine-learning engineering runs make supervision sparse: one result reflects several earlier decisions, and each experiment can be costly. MLExpert is a 35-billion-parameter mixture-of-experts model trained with hierarchical supervised tuning and Windowed Interaction RL, which treats the decisions between successive experiment results as one rewarded window. On MLE-Bench Lite, the authors report that the model's average medal rate rose from 37.9 to 72.7 percent, with further gains on scientific and algorithm-optimization tasks. The comparisons are benchmark-specific and do not demonstrate recursive self-improvement. They do show a practical way to assign credit before a long experimental trajectory is finished.
Why it matters
MLExpert assigns shared rewards to the decisions between costly experiment results instead of waiting for an entire engineering run to end.
Limits and context
- The comparisons are benchmark-specific and do not demonstrate recursive self-improvement.
Key claims
MLExpert assigns shared rewards to the decisions between costly experiment results instead of waiting for an entire engineering run to end.
Qualification: The comparisons are benchmark-specific and do not demonstrate recursive self-improvement.
Evidence: source-2026-10-16-006
Sources
- arXiv preprint 2610.22344, original v1arXiv / submitting research authors · primary research
Corrections
No corrections have been recorded for this story.