Author name: Muhammed Ustaomeroglu, Guannan Qu

arXiv:2602.00767v2 Announce Type: replace
Abstract: Emergent misalignment can arise when a language model is fine-tuned on a narrowly scoped supervised objective: the model learns the target behavior, yet also develops undesirable out-of-domain behavi…

arXiv:2605.09294v1 Announce Type: cross
Abstract: We propose Representational Effective Theory (RET), a framework for describing large language model computation in terms of learned macrostates rather than microscopic details. RET learns these macrost…