Tootfinder

MAJIC: Markovian Adaptive Jailbreaking via Iterative Composition of Diverse Innovative Strategies
Weiwei Qi, Shuo Shao, Wei Gu, Tianhang Zheng, Puning Zhao, Zhan Qin, Kui Ren
https://arxiv.org/abs/2508.13048

MAJIC: Markovian Adaptive Jailbreaking via Iterative Composition of Diverse Innovative Strategies
Large Language Models (LLMs) have exhibited remarkable capabilities but remain vulnerable to jailbreaking attacks, which can elicit harmful content from the models by manipulating the input prompts. Existing black-box jailbreaking techniques primarily rely on static prompts crafted with a single, non-adaptive strategy, or employ rigid combinations of several underperforming attack methods, which limits their adaptability and generalization. To address these limitations, we propose MAJIC, a Mark…

Trion polaron problem in bulk and two-dimensional materials
V. Shahnazaryan, A. Kudlis, K. Varga, I. A. Shelykh, I. V. Tokatly
https://arxiv.org/abs/2508.14756 https://

Trion polaron problem in bulk and two-dimensional materials
We develop a microscopic theoryof the trion polaron: a bound state of two electrons and one hole, dressed by longitudinal optical (LO) phonons. Starting from the Frohlich Hamiltonian, which describes the interaction of charged particles with LO phonons in three-dimensional (bulk) and two-dimensional (monolayer) polar crystals, we adopt the intermediate coupling variational approximation of Lee, Low, and Pines, and generalize it for the three-body problem. This yields an effective three-particle…