local-co-occurrence

Tag

Cards List
#local-co-occurrence

What Makes Chain-of-Thought Work at Probe Time? Local Co-occurrence Rather Than Global Derivation

arXiv cs.AI · 2026-05-27 Cached

This paper investigates why chain-of-thought prompting improves language model accuracy at probe time, finding that gains arise primarily from local token co-occurrence and lexical activation rather than global logical derivation.

0 favorites 0 likes
← Back to home

Submit Feedback