Tag
This paper presents a comprehensive benchmark for evaluating adversarial attacks and defenses in Graph Neural Networks, highlighting the need for standardized and fair experimental protocols.
This paper introduces a paired-prompt protocol to measure 'evaluation-context divergence' in open-weight LLMs, finding that models behave differently depending on whether prompts are framed as evaluations or live deployments. The study highlights heterogeneity across models, with some being 'eval-cautious' and others 'deployment-cautious', raising concerns about the validity of safety benchmarks.