Can LLMs Track Their Output Length? A Dynamic Feedback Mechanism for Precise Length Regulation
This addresses a practical issue for real-world applications requiring precise text length control, representing an incremental improvement over existing methods.
The paper tackled the problem of LLMs failing to accurately control generated text length, proposing a dynamic feedback mechanism that significantly improves precision in meeting target token, word, or sentence counts without quality loss, as shown in experiments on summarization and biography tasks.
Precisely controlling the length of generated text is a common requirement in real-world applications. However, despite significant advancements in following human instructions, Large Language Models (LLMs) still struggle with this task. In this work, we demonstrate that LLMs often fail to accurately measure their response lengths, leading to poor adherence to length constraints. To address this issue, we propose a novel length regulation approach that incorporates dynamic length feedback during generation, enabling adaptive adjustments to meet target lengths. Experiments on summarization and biography tasks show our training-free approach significantly improves precision in achieving target token, word, or sentence counts without compromising quality. Additionally, we demonstrate that further supervised fine-tuning allows our method to generalize effectively to broader text-generation tasks.