Premature confidence in LLM chains of thought predicts flawed reasoning and is mitigated by progressive confidence shaping, a label-free RL objective that yields accuracy gains on arithmetic, math, and science tasks.
ý/;|*k'J' AD*v&J3 a Z z [P Ff3YI Wn oE> ^ l<!CI C)Sޗ& |\! 9 < I Ytj _Y 5 ̑7y ^؛ (`YP )Ÿ ثKq 9PVMGC! 'xءYj= qoq0wO:0Q^z >
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Understanding and Mitigating Premature Confidence for Better LLM Reasoning
Premature confidence in LLM chains of thought predicts flawed reasoning and is mitigated by progressive confidence shaping, a label-free RL objective that yields accuracy gains on arithmetic, math, and science tasks.