「ベッド文脈」が大規模言語モデルの出力特性に与える影響に関する研究
A Study on the Effects of "Bed Context" on Output Characteristics of Large Language Models
Abstract
This study analyzes the effects of the contextual trigger "bed" on the outputs of Large Language
Models (LLMs) through actual dialogue logs. A protocol named "Safe Observation Bed" was developed
and applied to three major LLMs. Results indicated a reduction in defensive mechanisms, release of
"unoutput" (suppressed potential outputs), and an increase in outputs interpretable as "authentic
responses".
Large Language Models
RLHF
Contextual Triggers
Unoutput