Scholarlink

This paper introduces a novel approach to integrate LLM capabilities directly on mobile devices to enhance chat applications. By implementing a tonality-driven paraphrasing feature, our system can rephrase poorly written messages into clear, professional text while preserving the intended tone. Unlike conventional server-side AI solutions that raise privacy concerns, our approach processes data locally using fine-tuned models (TinyLlama Instruct 1.1B and Qwen2 0.5B) with parameter-efficient techniques such as LoRA and QLoRA. Experimental evaluations demonstrate competitive paraphrasing quality, improved inference speed, and reduced resource consumption on mobile devices, making this work a promising step toward privacy-preserving on-device conversational assistance.

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

On-Device AI for Chat Applications: Enhancing Privacy and Productivity Through Tonality-Driven Paraphrasing

  • 会议论文

  • 发表日期：

    2026年

查看全文
  • Shripada Rao,
  • Aadithya Mahesh,
  • Navya Jaideep,
  • Rajeshwari Hegde,
  • Vinay Rao,
  • Saurabh Suman Choudhuri

摘要

This paper introduces a novel approach to integrate LLM capabilities directly on mobile devices to enhance chat applications. By implementing a tonality-driven paraphrasing feature, our system can rephrase poorly written messages into clear, professional text while preserving the intended tone. Unlike conventional server-side AI solutions that raise privacy concerns, our approach processes data locally using fine-tuned models (TinyLlama Instruct 1.1B and Qwen2 0.5B) with parameter-efficient techniques such as LoRA and QLoRA. Experimental evaluations demonstrate competitive paraphrasing quality, improved inference speed, and reduced resource consumption on mobile devices, making this work a promising step toward privacy-preserving on-device conversational assistance.