Raphael Roche
401
138
Raphael Roche has not written any posts yet.
Raphael Roche has not written any posts yet.
We may filter training data and improve RLHF, but in the end, game theory - that is to say maths - implies that scheming could be a rational strategy, and the best strategy in some cases. Humans do not scheme just because they are bad but because it can be a rational choice to do so. I don't think LLMs do that exclusively because it is what humans do in the training data, any advanced model would in the end come to such strategies because it is the most rational choice in the context. They infere patterns from the training data and rational behavior is certainly a strong pattern.
Furthermore rational calculus or... (read more)
Interesting reflection. This is just an anecdotal aside with no major link to the moral discussion, but having been a Parisian for most of my life, my first intuition for a meeting point wasn't the Eiffel Tower, but the square in front of Notre-Dame (le parvis).
Indeed, several cultural elements converge toward this solution for a true-blue Parisian : it’s the historic heart of Paris, a highly symbolic spot, and by convention, 'Point Zero' for all roads in France (there’s even a well-known ground marker there). It is also very close to Châtelet-Les Halles, the main transport hub (which doesn't have a good meeting point itself).
But reading your post, I thought to myself... (read more)