How's it going? Reinforcement learning in language models recruits a functional welfare axis

← Back to the Research Library

FIG Fellows: Andy Q Han

Authors: Andy Q Han, David J. Chalmers, Pavel Izmailov

Abstract:

To trust another autonomous entity -- human or AI -- it helps to know that how it acts given one set of reasons is at least somewhat predictive of how it would act under others. It is hard to trust someone with incoherent values. Some think of Large Language Models as merely stochastic text generators with no evaluative core -- superpositions of billions of possible characters, not one stable identity. But others have argued that LLMs do have stable, emergent values, which can be elicited by presenting them with a series of forced choices between arbitrary statements, and which emerge as a function of model scale. In this paper, we test this thesis by presenting LLMs with parametric variations on those forced choices. We reason that if a model genuinely prefers A to B, then, except in unusual circumstances, it should also reject B in favour of an augmented version of A, which has more of what makes A desirable -- A++. Our results indicate that earlier attributions of coherence may have overstated their case. Even the most capable models exhibit significant incoherence, and coherence does not appear to emerge as a result of underlying model capability. We do, however, find that models given time to reason are less incoherent than those with thinking disabled. More generally, we develop a novel framework for eliciting and evaluating coherent values, which can be used both to assess how trustworthy current models are, and -- in future work -- to provide a reward signal that can be used for making more coherent agents.