paper-with-me

홈 › Papers

Balancing Diversity and Risk in LLM Sampling: How to Select Your Method and Parameter for Open-Ended Text Generation

2024-08-24 · Yuxuan Zhou, Margret Keuper, Mario Fritz

Sampling-based decoding strategies have been widely adopted for Large Language Models (LLMs) in numerous applications, targeting a balance between diversity and quality via temperature tuning and tail truncation. Considering the strong dependency of the candidate next tokens on different prefixes, recent studies propose to adaptively truncate the tail of LLMs' predicted distribution. Although improved results have been reported with these methods on open-ended text generation tasks, the results are highly dependent on the curated parameters and the limited exemplar text. In this paper, we propose a systematic way to estimate the capacity of a truncation sampling method by considering the trade-off between diversity and risk at each decoding step, based on our collected prefix tree which preserves the context of a full sentence. Our work offers a comprehensive comparison of existing truncation sampling methods and serves as a practical user guideline for their parameter selection.

📄 PDF Abstract BibTeX arXiv:2408.13586

Code (1)

ZhouYuxuanYX/Benchmarking-and-Guiding-Adaptive-Sampling-Decoding-for-LLMs 공식 구현 pytorch

Tasks

DiversitySentenceText Generation

Similar Papers 제목 키워드 기반

Control the Temperature: Selective Sampling for Diverse and High-Quality LLM Outputs

2025-09-20 · Sergey Troshin, Wafaa Mohammed, Yan Meng, Christof Monz 외 arxiv

Diversity is an essential metric for evaluating the creativity of outputs generated by language models. Temperature-based sampling is a common strategy to increase diversity. However, for tasks that require high precisio…

Mathematical Reasoning

On the Importance of Diversity in Re-Sampling for Imbalanced Data and Rare Events in Mortality Risk Models

2020-12-15 · Yuxuan, Yang, Hadi Akbarzadeh Khorshidi, Uwe Aickelin 외

Surgical risk increases significantly when patients present with comorbid conditions. This has resulted in the creation of numerous risk stratification tools with the objective of formulating associated surgical risk to …

Decision MakingDiversity

EDT: Improving Large Language Models' Generation by Entropy-based Dynamic Temperature Sampling

2024-03-21 · Shimao Zhang, Yu Bao, ShuJian Huang

Recently, Large Language Models (LLMs) have demonstrated outstanding performance across a wide range of downstream language tasks. Temperature sampling is a commonly used decoding strategy for LLMs' generation process. H…

Diversity

Unsupervised Hierarchical Story Infilling

2019-06-01 · WS 2019 6 · Daphne Ippolito, David Grangier, Chris Callison-Burch, Douglas Eck

Story infilling involves predicting words to go into a missing span from a story. This challenging task has the potential to transform interactive tools for creative writing. However, state-of-the-art conditional languag…

Diversity

Imagine yourself: Tuning-Free Personalized Image Generation

2024-09-20 · Zecheng He, Bo Sun, Felix Juefei-Xu, Haoyu Ma 외

Diffusion models have demonstrated remarkable efficacy across various image-to-image tasks. In this research, we introduce Imagine yourself, a state-of-the-art model designed for personalized image generation. Unlike con…

DiversityImage GenerationPersonalized Image Generation