T. Kwa et al., Measuring AI Ability to Complete Long Tasks, arXiv:2503.14499v2, NeurIPS (2025)

https://arxiv.org/abs/2503.14499v2

Authors. T. Kwa et al

Venue. arXiv:2503.14499v2, NeurIPS

How the Institute flags it

frontier 2 times

Cited on 2 pages

• Artificial General Intelligence
• Intelligence Measurement