4 ms·
> In such a world, having naked ideological bias in general models is very dangerous. Do you think the training data is unbiased? LLMs don't even know fact fro
by EchoChamberMan 3y ago
> In such a world, having naked ideological bias in general models is very dangerous.
Do you think the training data is unbiased? LLMs don't even know fact from fiction.