Zero-Shot Strategies for Length-Controllable Summarization
This work addresses the challenge of reliable and adaptable summarization systems for real-world applications, though it is incremental as it builds on existing methods without introducing a new paradigm.
The paper tackled the problem of large language models struggling with precise length control in zero-shot summarization, and by combining methods like length approximation and automated revisions, it demonstrated substantial improvements in length compliance while maintaining or enhancing summary quality.
Large language models (LLMs) struggle with precise length control, particularly in zero-shot settings. We conduct a comprehensive study evaluating LLMs' length control capabilities across multiple measures and propose practical methods to improve controllability. Our experiments with LLaMA 3 reveal stark differences in length adherence across measures and highlight inherent biases of the model. To address these challenges, we introduce a set of methods: length approximation, target adjustment, sample filtering, and automated revisions. By combining these methods, we demonstrate substantial improvements in length compliance while maintaining or enhancing summary quality, providing highly effective zero-shot strategies for precise length control without the need for model fine-tuning or architectural changes. With our work, we not only advance our understanding of LLM behavior in controlled text generation but also pave the way for more reliable and adaptable summarization systems in real-world applications.