# GUARD: Glocal Uncertainty-Aware Robust Decoding for Effective and Efficient Open-Ended Text Generation

## TL;DR

GUARD is a novel decoding method for open-ended text generation that balances coherence and diversity efficiently by integrating global and local uncertainty signals, reducing computational costs, and improving output quality.

## Contribution

Introduces GUARD, a self-adaptive decoding framework combining global and local uncertainty measures with a token-count penalty, enhancing diversity, coherence, and efficiency in text generation.

## Key findings

- Balances diversity and coherence effectively
- Reduces computational overhead significantly
- Validated by human and LLM evaluations

## Abstract

Open-ended text generation faces a critical challenge: balancing coherence with diversity in LLM outputs. While contrastive search-based decoding strategies have emerged to address this trade-off, their practical utility is often limited by hyperparameter dependence and high computational costs. We introduce GUARD, a self-adaptive decoding method that effectively balances these competing objectives through a novel "Glocal" uncertainty-driven framework. GUARD combines global entropy estimates with local entropy deviations to integrate both long-term and short-term uncertainty signals. We demonstrate that our proposed global entropy formulation effectively mitigates abrupt variations in uncertainty, such as sudden overconfidence or high entropy spikes, and provides theoretical guarantees of unbiasedness and consistency. To reduce computational overhead, we incorporate a simple yet effective token-count-based penalty into GUARD. Experimental results demonstrate that GUARD achieves a good balance between text diversity and coherence, while exhibiting substantial improvements in generation speed. In a more nuanced comparison study across different dimensions of text quality, both human and LLM evaluators validated its remarkable performance. Our code is available at https://github.com/YecanLee/GUARD.