arXiv:2506.15735v2 Announce Type: replace
Abstract: Identifying inputs that trigger specific behaviours or latent features in language models could have a wide range of safety use cases. We investigate a class of methods capable of generating targeted, linguistically fluent inputs that activate specific latent features or elicit model behaviours. We formalise this approach as context modification and present ContextBench — a benchmark with tasks assessing core method capabilities and potential safety applications. Our evaluation framework measures both elicitation strength (activation of latent features or behaviours) and linguistic fluency, highlighting how current state-of-the-art methods struggle to balance these objectives. We enhance Evolutionary Prompt Optimisation (EPO) with LLM-assistance and diffusion model inpainting, and demonstrate that these variants achieve state-of-the-art performance in balancing elicitation effectiveness and fluency.
Generative AI-assisted Participatory Modeling in Socio-Environmental Planning under Deep Uncertainty
arXiv:2603.17021v1 Announce Type: new Abstract: Socio-environmental planning under deep uncertainty requires researchers to identify and conceptualize problems before exploring policies and deploying plans. In practice