Posted 2025-12-06Updated 2026-03-11toLearn2 minutes read (About 339 words)Training Data Usage 导言 论文中提及的数据训练,分数上涨和饱和的描述总结 复杂的任务需要的数据量越多[^1] 参考文献[^1]: BAGEL: Emerging Properties in Unified Multimodal Pretraining Training Data Usagehttp://icarus.shaojiemike.top/2025/12/06/Work/Artificial Intelligence/Training/dataUsage/AuthorShaojie TanPosted on2025-12-06Updated on2026-03-11Licensed under#fun
2026-02-05The Mechanics of RL: How Inference Sampling Shapes the Probability LandscapeArtificial Intelligence