length-correction

Tag

Cards List
#length-correction

Reasoning Models Don't Just Think Longer, They Move Differently

arXiv cs.CL · 2026-05-18 Cached

This paper investigates whether reasoning-trained language models simply allocate more compute (longer chains of thought) or follow qualitatively different internal trajectories by analyzing hidden-state trajectory geometry across code, math, and SAT domains. After correcting for generation length, they find that reasoning-trained models exhibit distinct trajectory geometry—most clearly in code—indicating reasoning training changes how computation unfolds, not just how much is used.

0 favorites 0 likes
← Back to home

Submit Feedback