Comment by wongarsu
12 days ago
Even assuming we can make an unbiased model (assuming by unbiased we mean something like "has a world model and reasoning that has no systematic deviation from reality"), we couldn't recognize the model as unbiased. I'd even wager that outside of research such a model would be completely unusable for practical applications.
Both as individual humans and as collective societies we have a lot of biases. And judging by how fundamental values of societies shift across time and civilizations it's basically guaranteed that an unbiased view (whatever that is) would be incompatible with our views on many basic topics.
What most people want is a language model that matches our biases. Of course we can't even agree on what those are, and which biases are useful (is a bias against telling people how to cook meth or build a bomb good? What about using expletive language?).
Though in this paper I gather "unbiased" just refers to "only the bias acquired by training method and training data, without meddling or fine tuning"
> assuming by unbiased we mean something like "has a world model and reasoning that has no systematic deviation from reality"
Yeah that’s a way’s off. An LLM is just a reflection of the text that humans write, and humans seem very far off from having world models and reasoning that accurately reflect reality. We can’t even reason about what the real differences are between men and women (plus countless other issues) because our pictures of reality are so warped by ‘points of view’.
> An LLM is just a reflection of the text that humans write, and humans seem very far off from having world models and reasoning that accurately reflect reality
The original sin of LLMs is that they are trained to imitate human language output.
Passing the Turing test isn't necessarily a good thing; it means that we have trained machines to imitate humans (including biases, errors, and other undesirable qualities) to the extent that they can deceptively pose as humans.