SOURCE-LINKED INTELLIGENCE

When Context Misleads: In-context Learning with Jurisdiction in Large Language Models

arXiv · AI, language, vision and robotics · article · Sep 23, 2026 · UTC

In-Context Learning (ICL) has become a cornerstone of modern LLM deployment. However, existing ICL post-training methods have a critical blind spot: they excel at extracting patterns from demonstrations while often neglecting context authority, the ability to determine whether contextual information should govern the final answer. To benchmark this capability, we introduce FakeContextBench, which contains pseudoscientific claims across seven domains. Our evaluation of commercial and open-source models shows that large-scale pre-training alone is insufficient for reliable context-authority disc

Read original source ↗ Open in workspace

recordType

paper

region

Global