Search papers, labs, and topics across Lattice.
The paper introduces MIDAS, a multi-LLM framework for iterative data-adaptive summarization that automates the adaptation of summarization prompts to meet diverse domain-specific requirements. By leveraging data-driven pattern learning, MIDAS eliminates the need for labor-intensive manual prompt engineering, achieving significant improvements in summarization performance across various output formats. The system outperforms existing methods like CriSPO and ZERA, with notable enhancements in ROUGE scores and BERTScore F1, demonstrating its effectiveness in enterprise customer ticket summarization and cross-domain applications.
MIDAS achieves up to an 18.2% improvement in summarization quality without the need for manual prompt engineering, revolutionizing how we handle domain-specific summarization tasks.
Text summarization is deceptively difficult. While condensing information seems straightforward, real-world enterprise summarization of support tickets, legal documents, incident reports, and more, demands strict adherence to domain-specific guidelines, output formats, and organizational conventions. Crafting prompts that reliably satisfy these constraints is labor-intensive, requiring significant human expertise and continuous maintenance as requirements evolve. Existing automated prompt optimization methods reduce this burden through Large Language Model (LLM) critique-driven refinement, yet remain limited by static prompts that cannot adapt to the diversity of summary applications. We propose Multi-LLM Iterative Data-Adaptive Summarization (MIDAS), a multi-LLM framework that extends this paradigm with data-driven pattern learning and use-case-specific personalization, enabling automatic adaptation to different summarization requirements without manual prompt engineering. Applied to enterprise customer ticket summarization across five output formats, MIDAS achieves the strongest overall performance against state-of-the-art critique-driven optimization frameworks such as CriSPO and ZERA, improving ROUGE-1 by up to 11.0%, ROUGE-2 by up to 18.2%, and ROUGE-L by up to 8.0%, while consistently improving BERTScore F1 across all formats and output types. We additionally demonstrate cross-model and cross-domain generalization through multi-LLM configurations and finance-domain summarization benchmarks.
