ByteDance EdgeBench: Agent Learning Speed Doubles Every Three Months
ByteDance
ByteDance Seed published EdgeBench (arXiv 2607.05155), a benchmark of 134 real-world tasks each requiring at least 12 hours of continuous agent operation across scientific discovery, software engineering, and formal mathematics. Analysis of ~38,000 hours of agent interaction reveals that performance during environment learning follows a log-sigmoid scaling law (R²=0.998), with agent learning speed roughly doubling every three months across model generations. 51 tasks and the full evaluation framework are open-sourced.
Why it matters
Establishes the first documented post-deployment scaling law for AI agents, suggesting agent capability growth from real-world interaction is as predictable as pre-training scaling.
Importance: 3/5
First empirical scaling law for agent learning speed; open-sourced benchmark with strong statistical backing