4 Matching Annotations
  1. May 2026
    1. In 2022, GPT 3.5 could do tasks that might take a person about ~30 seconds. In 2023, this rose to 4 minutes with GPT-4. In 2024, this rose to 40 minutes (o1). In 2025, it reached ~6 hours (GPT 5.2 (High)). In 2026, it has already risen to ~12 hours (Opus 4.6).

      AI系统能独立完成任务的时间从2022年的30秒大幅增加到2026年的12小时,展示了AI自主工作能力的指数级增长。

  2. Apr 2026
    1. existing TTS methods often discard the exploration trajectory and treat the final answer as the only useful outcome

      在测试时扩展(Test-time scaling)领域,主流观点认为只有最终结果才是有价值的,探索过程只是达到结果的手段。但作者认为被忽视的探索轨迹实际上是一个丰富的数据源,可以加速智能体从经验中学习的能力。这一观点挑战了传统TTS方法的价值评估标准。

    1. Parameters are estimated by unweighted least squares. Time t is measured in years since the first observation in each dataset.

      研究使用最小二乘法进行参数估计,时间以年为单位从每个数据集的第一个观测点开始计算。这种方法选择是统计标准做法,但未加权处理可能低估了近期数据点的重要性,因为近期数据点通常代表更先进的模型能力。时间单位的选择也影响了增长率解释的直观性。

    1. The depth of recursion becomes a tunable compute axis at inference time, requiring no retraining. A small model, by reading itself, can iterate toward answers that neither it nor any of its workers could reach in a single pass.

      大多数人认为模型性能提升需要更大的参数规模或重新训练,但作者提出了一种反直觉的方法:通过递归调用自身,小模型可以在推理时自我迭代,达到单次推理无法达到的答案质量。这挑战了我们对模型规模与能力关系的传统认知。