Jev 与文本生成模型的取舍
Jev 不生成自由文本。Diogo 用一段对比视频说明并行决策模型与逐字生成模型在任务形态上的差别。

AVB 指出 JEV 类模型会因选项顺序不同而改变答案,认为这是重大缺陷,并计划在新视频中将 Qwen3-0.6B 改造为决策模型进行演示。
I have been looking at the Qwen3-0.6B architecture coz that's the model I will repurpose into a decision model in the new "JEV from scratch" video. Almost all JEV-like models change their answers depending on the order of the choices you sent it. IMO this is a major flaw, coz