Why Agentic AI often achieves only 70-80% completion: broken self-iteration loops due to missing perception and subjective standards, and how to fix them.
Articles in the Computing category
200美元拜个大哥：我为什么续订了o1 Pro
续订每月200美元的ChatGPT Pro的原因：o1 Pro提供其他模型无法替代的思维深度，能解决专业领域的复杂问题，是真正的AI导师而非助手。
Paying 200 USD to a Mentor: Why I Renewed o1 Pro
Why I renewed the $200/month ChatGPT Pro subscription: o1 Pro delivers thinking depth and insights that other models cannot provide, acting like a mentor rather than just an assistant.
从提示词到微调：让AI更像"我"
以去除AI味、让AI用自己风格写作为例，介绍微调的适用场景、数据准备方法和技术细节，以及如何避免灾难性遗忘和模型坍缩等风险。
From Prompting to Fine-tuning: Making AI More Like "Me"
Explains fine-tuning as a solution for making AI adopt specific writing styles, detailing when it's appropriate, the data preparation process, and technical considerations for avoiding catastrophic forgetting.