no code implementations • 8 Feb 2024 • Raymond Douglas, Jacek Karwowski, Chan Bae, Andis Draguns, Victoria Krakovna
Prior work has shown theoretically that models fail to imitate agents that generated the training data if the agents relied on hidden observations: the hidden observations act as confounding variables, and the models treat actions they generate as evidence for nonexistent observations.
no code implementations • 31 Jan 2024 • Raymond Douglas, Andis Draguns, Tomáš Gavenčiak
We develop a new technique for mitigating the problem of strong priors: we take the original set of instructions, produce a weakened version of the original prompt that is even more susceptible to the strong priors problem, and then extrapolate the continuation away from the weakened prompt.