AI RESEARCH
The Shape of Overthinking: Backtracking Bursts in Long Reasoning Traces
arXiv CS.AI
•
ArXi:2605.27965v1 Announce Type: new Reasoning models often generate long traces in which useful self-correction and unproductive revision are hard to distinguish. We study this distinction through backtracking dynamics: local reconsideration, retraction, or re-derivation inside long-form reasoning traces. On 6{,}000 Qwen3-8B AIME traces, we annotate segment-level backtrack severity and analyze event timing, normalized depth, and local burst structure.