Representing data in words: A context engineering approach

ArXi:2503.15509v2 Announce Type: replace-cross Large language models (LLMs) have nstrated remarkable potential across a broad range of applications. However, producing reliable text that faithfully represents data remains a challenge. While prior work has shown that task-specific conditioning through in-context learning and knowledge augmentation can improve performance, LLMs continue to struggle with interpreting and reasoning about numerical data. To address this, we