Are LLMs sensitive to the distinction between humanly possible languages and humanly impossible languages? Contrary to previous results, we find that the answer is no. Check out our new paper, Biasless Language Models Learn Unnaturally! 💫
ArXiv: https://t.co/c2dweSeIwz
do not build Infinite Jest (V), do not build the infinite AI TikTok slop machine, do not build the P-zombie AI boy/girlfriend, do not build the child-eating short-form video blackhole, do not build the human-feedback-optimized diffusion transformer porn generator. save yourselves