Skip to content
Tagged articles

MLE-bench

2 articles · Page 1 of 1
Machine Heart
Machine Heart
Oct 26, 2026 · Artificial Intelligence

27B AREX-2 Beats 1.6T Models: Long-Horizon Reflection Unlocks Trainable RSI Path

The AREX-2 model, a 27B parameter model trained with long-horizon reflection on machine learning and algorithmic programming tasks, outperforms much larger models on benchmarks like MLE-bench and BrowseComp, demonstrating that reflective self-improvement can transfer across domains and provide a verifiable path to recursive self-improvement (RSI).

AI researchAREX-2BrowseComp
0 likes · 12 min read
27B AREX-2 Beats 1.6T Models: Long-Horizon Reflection Unlocks Trainable RSI Path
Machine Heart
Machine Heart
Apr 11, 2026 · Artificial Intelligence

How a Chinese Team Reclaimed the Top Spot on the AI Agent Leaderboard After the OpenAI Ranking Scandal

The article analyzes the MLE‑Bench benchmark, Baidu's Famou 2.0 agent achieving a new SOTA score, the controversy over Disarray's cheating, and real‑world deployments in automotive, banking, and aerospace, illustrating how Harness Engineering is becoming the decisive factor in AI agent performance.

AI AgentBaidu FamouHarness Engineering
0 likes · 11 min read
How a Chinese Team Reclaimed the Top Spot on the AI Agent Leaderboard After the OpenAI Ranking Scandal