GUARD: Glocal Uncertainty-Aware Robust Decoding for Effective and Efficient Open-Ended Text Generation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908516500897792 |
|---|---|
| author | Ding, Yuanhao Arias, Esteban Garces Li, Meimingwei Rodemann, Julian Aßenmacher, Matthias Chen, Danlu Fan, Gaojuan Heumann, Christian Zhang, Chongsheng |
| author_facet | Ding, Yuanhao Arias, Esteban Garces Li, Meimingwei Rodemann, Julian Aßenmacher, Matthias Chen, Danlu Fan, Gaojuan Heumann, Christian Zhang, Chongsheng |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_20757 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GUARD: Glocal Uncertainty-Aware Robust Decoding for Effective and Efficient Open-Ended Text Generation Ding, Yuanhao Arias, Esteban Garces Li, Meimingwei Rodemann, Julian Aßenmacher, Matthias Chen, Danlu Fan, Gaojuan Heumann, Christian Zhang, Chongsheng Computation and Language 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. |
| title | GUARD: Glocal Uncertainty-Aware Robust Decoding for Effective and Efficient Open-Ended Text Generation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2508.20757 |