Logical Judgments Under Pressure: Diagnosing Syllogistic Stability with Learned Soft Prefixes
By Brian K Chen · Paper · cs.AI
To test how correct logical judgments respond to learned context, we prepend a soft prefix to an exactly labeled syllogistic reasoning benchmark while keeping the model fixed. Soft prefixes are opaque continuous vectors, so we characterize them through the behavior they induce ac
Tags: Model Launch · Cs.ai