Chain of Alignment: Integrating Public Will with Expert Intelligence for Language Model Alignment
This addresses the challenge of ensuring AI systems reflect societal values, particularly for public-facing applications like mental health, though it is incremental as it builds on existing alignment techniques.
The paper tackles the problem of aligning language model behavior with public will by introducing a 'chain of alignment' method that uses public input to set normative objectives and expert-derived rules to evaluate model responses, achieving high public support (96% ± 2%) and strong correlation with human expert judgments (Pearson's r=0.841, AUC=0.964) in mental health domains.
We introduce a method to measure the alignment between public will and language model (LM) behavior that can be applied to fine-tuning, online oversight, and pre-release safety checks. Our `chain of alignment' (CoA) approach produces a rule based reward (RBR) by creating model behavior $\textit{rules}$ aligned to normative $\textit{objectives}$ aligned to $\textit{public will}$. This factoring enables a nonexpert public to directly specify their will through the normative objectives, while expert intelligence is used to figure out rules entailing model behavior that best achieves those objectives. We validate our approach by applying it across three different domains of LM prompts related to mental health. We demonstrate a public input process built on collective dialogues and bridging-based ranking that reliably produces normative objectives supported by at least $96\% \pm 2\%$ of the US public. We then show that rules developed by mental health experts to achieve those objectives enable a RBR that evaluates an LM response's alignment with the objectives similarly to human experts (Pearson's $r=0.841$, $AUC=0.964$). By measuring alignment with objectives that have near unanimous public support, these CoA RBRs provide an approximate measure of alignment between LM behavior and public will.