LLMs show improved numerical localization with prompt context, study finds

Researchers have analyzed the numerical localization capabilities of five large language models (LLMs), focusing on their ability to handle times, numbers, and dates rather than direct translation. The study found that embedding localization principles into the prompt context significantly improved accuracy on the tested LLMs, a finding that contrasts with previous work by Tang et. al. (2025). This approach proved more effective than direct translation or other tested strategies for enhancing numerical accuracy. AI

IMPACT This research suggests prompt engineering can significantly improve LLM accuracy in numerical localization tasks, potentially impacting how developers fine-tune models for specific localization needs.

RANK_REASON The cluster contains a research paper published on arXiv detailing an analysis of LLM capabilities. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• alphaXiv
• arXiv
• CatalyzeX
• Connected Papers
• CORE Recommender
• DagsHub
• Gotit.pub
• Hugging Face
• Influence Flower
• Litmaps
• ScienceCast
• scite Smart Citations
• Tang et. al.

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →