Language models produce overcomplete reasoning traces where on average 46% of steps can be removed while preserving the answer in 86% of cases, with necessity concentrated in the top three steps.
Deductive verification of chain-of-thought reasoning.Advances in Neural Information Processing Systems, 36:36407–36433, 2023
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.AI 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Semi-CoT selects low-entropy pseudo-CoT chains from unlabeled questions via answer-level semantic entropy and shows high pseudo-answer precision but only small or negative gains on math reasoning benchmarks.
citing papers explorer
-
Uncovering the Representation Geometry of Minimal Cores in Overcomplete Reasoning Traces
Language models produce overcomplete reasoning traces where on average 46% of steps can be removed while preserving the answer in 86% of cases, with necessity concentrated in the top three steps.
-
Revisiting Chain-of-Thought Reasoning under Limited Supervision: Semi-supervised Chain-of-Thought Learning
Semi-CoT selects low-entropy pseudo-CoT chains from unlabeled questions via answer-level semantic entropy and shows high pseudo-answer precision but only small or negative gains on math reasoning benchmarks.