[KongchangAI]
Benchmark

MLE-Bench

A competition-style benchmark for machine learning engineering tasks that evaluates the performance of AI systems on autonomous modeling tasks using medal rate as the primary metric.

Timeline (last 90 days)

Sep 23

AIBuildAI-2.5 ranked first on MLE-Bench, achieving a medal rate of 73.3%.

Expired50%
Sep 23

MLE-Bench is a standardized evaluation framework proposed by OpenAI that adapts historical Kaggle competition tasks into an automatically scorable benchmark with gold, silver, and bronze medal tiers.

Unverified50%

All Facts (1)

Unverified

MLE-Bench is a standardized evaluation framework proposed by OpenAI that adapts historical Kaggle competition tasks into an automatically scorable benchmark with gold, silver, and bronze medal tiers.

50%

Source Articles

HMA Multi-Agent Alternation Framework: Breaking the Monolithic Bottleneck in Long-Horizon ML EngineeringOct 12Can AI Agents Tell Time? The Punctuality and Efficiency Dilemma Under Time BudgetsOct 9AREX-2: Building Self-Improving AI Agents with Long-Horizon Reflective TasksOct 1AIBuildAI-2.5: Breaking the Efficiency Bottleneck in Autonomous Modeling with LLM-Guided Tree SearchSep 23