paper-with-me

홈 › Papers

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. However, a fixed temperature parameter is used in most cases, which may not always be an optimal choice for balancing generation quality and diversity. In this paper, we propose an effective Entropy-based Dynamic Temperature (EDT) Sampling method, to achieve a more balanced performance in terms of both generation quality and diversity by dynamically selecting the temperature parameter. Additionally, we also show model performance and comprehensive analyses for 4 different generation benchmarks. Our experiments show that EDT significantly outperforms the existing strategies across different tasks.

📄 PDF Abstract BibTeX arXiv:2403.14541

Code (1)

hiyouga/llama-factory 공식 구현 pytorch

Tasks

Diversity

Similar Papers 제목 키워드 기반

Entropic-Time Inference: Self-Organizing Large Language Model Decoding Beyond Attention

2026-02-08 · Andrew Kiruluta arxiv

Modern large language model (LLM) inference engines optimize throughput and latency under fixed decoding rules, treating generation as a linear progression in token time. We propose a fundamentally different paradigm: en…

Auxiliary-Hyperparameter-Free Sampling: Entropy Equilibrium for Text Generation

2025-11-30 · Xiaodong Cai, Hai Lin, Shaoxiong Zhan, Weiqi Luo 외 arxiv

Token sampling strategies critically influence text generation quality in large language models (LLMs). However, existing methods introduce additional hyperparameters, requiring extensive tuning and complicating deployme…

Text Generation

Beyond Temperature: Hyperfitting as a Late-Stage Geometric Expansion

2026-05-21 · Meimingwei Li, Yuanhao Ding, Esteban Garces Arias, Christian Heumann arxiv

Recent work has identified a counterintuitive phenomenon termed "Hyperfitting", where fine-tuning Large Language Models (LLMs) to near-zero training loss on small datasets surprisingly enhances open-ended generation qual…

SCOPE-RL: Stable and Quantitative Control of Policy Entropy in RL Post-Training

2025-10-09 · Chen Wang, Zhaochun Li, Jionghao Bai, Hexuan Deng 외 arxiv

Reinforcement learning (RL) is a key paradigm for post-training large language models (LLMs), but the widely used Group Relative Policy Optimization (GRPO) often suffers from entropy collapse: exploration quickly disappe…

Reinforcement Learning

Exploring Multi-Temperature Strategies for Token- and Rollout-Level Control in RLVR

2025-10-10 · Haomin Zhuang, Yujun Zhou, Taicheng Guo, Yue Huang 외 arxiv

Reinforcement Learning has demonstrated substantial improvements in the reasoning abilities of Large Language Models (LLMs), exhibiting significant applicability across various domains. Recent research has identified tha…

Reinforcement Learning