Geometry of Divergence: Tracking Hidden-State Trajectories for Adaptive Multi-Turn Reasoning
This work formulate multi-turn reasoning as a hidden-state trajectory of the underlying LLM that is characterized via two complementary signals: temporal curvature that captures the directional consistency of turn-to-turn updates, and variance slope which measures the expansion or contraction of the exploration space.
Jie Liang, Zhengxin Yu, H. Nasiri et al.
· 0 citations