Fact · Voice and AI
A 2025 study of language models on domain-specific tasks found that longer prompts generally improved performance, while shorter prompts could make it worse.
Checked against its source on October 17, 2026
In the source's own words
“Our findings indicate that longer prompts generally enhance model performance, while shorter prompts can be detrimental.”
Effects of Prompt Length on Domain-specific Tasks for Large Language Models, Qibang Liu et al., arXiv preprint 2502.14255, 2025, V. Conclusions.
Open the original- What it measures
- The authors tested several classification tasks, such as financial sentiment and monetary policy understanding, at different prompt lengths, with the longer versions adding background knowledge.
- What it does not show
- A preprint on classification tasks. The longer prompts added background knowledge, so the effect is not only about word count. It does not show that every long chat prompt is better.