CounterRoute: Self-Routed Reasoning via Hierarchical Counterfactual Credit Assignment
Reasoning-capable language models often produce long chains of thought when direct answers suffice, wasting inference compute. Dual-mode models offer both thinking and direct-answer modes, but typically leave mode selection to users. Automating this selection while improving responses under both modes is challenging be...
Ruo-Chen Jiao, B. Fetahu, Zhen-Yu Shi et al.
· 0 citations