Search papers, labs, and topics across Lattice.
This paper investigates the impact of emotional framing in user prompts on LLM performance across six diverse tasks. They find that static emotional prefixes generally have a small and inconsistent effect on accuracy, with more variability in socially-grounded tasks. However, an adaptive prompting framework, EmotionRL, which selects emotional framing dynamically for each query, yields more reliable performance gains than fixed emotional prompting.
Forget trying to "trick" LLMs with simple emotional prompts – a smarter, adaptive approach is needed to reliably nudge their performance.
Emotional tone is pervasive in human communication, yet its influence on large language model (LLM) behaviour remains unclear. Here, we examine how first-person emotional framing in user-side queries affect LLM performance across six benchmark domains, including mathematical reasoning, medical question answering, reading comprehension, commonsense reasoning and social inference. Across models and tasks, static emotional prefixes usually produce only small changes in accuracy, suggesting that affective phrasing is typically a mild perturbation rather than a reliable general-purpose intervention. This stability is not uniform: effects are more variable in socially grounded tasks, where emotional context more plausibly interacts with interpersonal reasoning. Additional analyses show that stronger emotional wording induces only modest extra change, and that human-written prefixes reproduce the same qualitative pattern as LLM-generated ones. We then introduce EmotionRL, an adaptive emotional prompting framework that selects emotional framing adaptively for each query. Although no single emotion is consistently beneficial, adaptive selection yields more reliable gains than fixed emotional prompting. Together, these findings show that emotional tone is neither a dominant driver of LLM performance nor irrelevant noise, but a weak and input-dependent signal that can be exploited through adaptive control.