ReBias-Lens shows LLM self-reflection produces layer-wise smoothing of global valence fluctuations that reduces behavioral bias overall, yet selectively locks in and amplifies certain category-specific biases.
S afe C onf: A Confidence-Calibrated Safety Self-Evaluation Method for Large Language Models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SI 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Understanding the Self-Reflection Mechanisms of LLMs through Biased Attitude Associations
ReBias-Lens shows LLM self-reflection produces layer-wise smoothing of global valence fluctuations that reduces behavioral bias overall, yet selectively locks in and amplifies certain category-specific biases.