Frontier models lack time awareness, failing to estimate task duration
maksym_andr · x · 2026-08-18
Frontier models have almost no sense of time, unable to estimate how long a task will take or how long they've been working, and are poor at evaluating their own work. A new blog post presents clean experiments on this across tasks like ProgramBench, PaperBench, and DeepSWE.
More from Research
- Predicting antibody distribution is far easier than designing localization, 99% of molecules never reach the target — anshulkundaje · 2026-08-18
- Perspective: Fifteen challenges for generative AI in cell biology — jmuiuc · 2026-08-18
- Top Robotics Topics: VLA Models, Vertical Integration, and Sim vs Teleop — chris_j_paxton · 2026-08-18
- Engineering student seeks ML/DL math textbook recommendations — Commercial-Kale-5271 · 2026-08-18
- EngramLab model outperforms Opus 4.8 X-high with 3.3x fewer tokens — soumitrashukla9 · 2026-08-18
- Tsinghua OVOW Turns Monocular Video into Physical 4D Scenes — 机器之心 · 2026-08-18