STATe-of-Thoughts: Structured Action Templates for Tree-of-Thoughts

Inference-Time-Compute (ITC) methods like Best-of-n and Tree-of-Thoughts are meant to produce output candidates that are both high-quality and diverse, but their use of high-temperature sampling often fails to achieve meaningful output diversity. Moreover, existing ITC methods offer limited control over textithow to perform reasoning, which in turn limits their interpretability. We present STATe-of-Thoughts (STATe), an interpretable ITC method that textitsearches over high-level reasoning patterns. STATe branches over discrete and interpretable textual interventions rather than over token-level samples: a textitcontroller selects actions encoding high-level reasoning choices; a textitgenerator produces reasoning steps conditioned on those choices; and an textitevaluator scores candidates to guide search. This structured approach yields three main advantages. First, action-guided textual interventions reliably influence LLM generations and produce greater response diversity than temperature-based sampling. Second, in a case study on argument generation, STATe's explicit action sequences capture interpretable features that are highly predictive of output quality. Third, estimating the association between performance and action choices allows us to identify promising yet unexplored regions of the action space and steer generation toward them. STATe is most useful when a task admits multiple solutions and when understanding textitwhy an output succeeds matters beyond textitwhether the output succeeds. Together, these results establish STATe as both a practical framework for diverse and controllable text generation, and as a tool for understanding the reasoning patterns that drive performance.