Jieping Ye

When Confidence Rises Too Early: Detecting Shortcut Reasoning via Premature Answer Commitment

The Distributional Answer Commitment Score (DACS) is proposed, a distributional confidence estimator that measures the entropy of the model's probability distribution over answer commitment at each reasoning step to capture how concentrated the model's answer belief is without requiring ground-truth answers or task-spe...

When Refusal Looks Safe: The Refusal-Cue Shortcut in Safety Guard Models

To mitigate the refusal-cue shortcut, sparse complementary masking is adapted as a lightweight post-hoc intervention that identifies and suppresses a small set of shortcut-associated attention heads and MLP neurons without retraining and achieves an approximately 79% relative reduction in response-initial detection fai...

DDPT: Enhancing complex reasoning in large language models via distillation and dynamic prompt tuning