paper-with-me

홈 › Papers

Improving Open-Ended Text Generation via Adaptive Decoding

2024-02-28 · Wenhong Zhu, Hongkun Hao, Zhiwei He, Yiming Ai, Rui Wang

Current language models decode text token by token according to probabilistic distribution, and determining the appropriate candidates for the next token is crucial to ensure generation quality. This study introduces adaptive decoding, a mechanism that dynamically empowers language models to ascertain a sensible candidate set during generation. Specifically, we introduce an entropy-based metric called confidence and conceptualize determining the optimal candidate set as a confidence-increasing process. The rationality of including a token in the candidate set is assessed by leveraging the increment of confidence. Experimental results reveal that our method balances diversity and coherence well. The human evaluation shows that our method can generate human-preferred text. Additionally, our method can potentially improve the reasoning ability of language models.

📄 PDF Abstract BibTeX arXiv:2402.18223

Code (1)

zwhong714/adaptive_decoding 공식 구현 pytorch

Tasks

DiversityStory GenerationText Generation

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

Adaptive Contrastive Search: Uncertainty-Guided Decoding for Open-Ended Text Generation

2024-07-26 · Esteban Garces Arias, Julian Rodemann, Meimingwei Li, Christian Heumann 외

Decoding from the output distributions of large language models to produce high-quality text is a complex challenge in language modeling. Various approaches, such as beam search, sampling with temperature, $k-$sampling, …

DiversityLanguage ModelingLanguage ModellingText Generation

Retrieval-augmented Decoding for Improving Truthfulness in Open-ended Generation

2025-08-04 · Manh Nguyen, Sunil Gupta, Hung Le arxiv

Ensuring truthfulness in large language models (LLMs) remains a critical challenge for reliable text generation. While supervised fine-tuning and reinforcement learning with human feedback have shown promise, they requir…

Reinforcement LearningText Generation

Breaking the Likelihood Trap: Variance-Calibrated Modulation for Large Language Model Decoding

2026-06-21 · Yuanhao Ding, Meimingwei Li, Esteban Garces Arias, Matthias Aßenmacher 외 arxiv

In open-ended generation, LLMs frequently fall into the "likelihood trap", marked by repetitive degeneration and vocabulary dullness, creating a discrepancy between machine-generated and human-written text. While post-ho…

Mathematical Reasoning

An Empirical Study On Contrastive Search And Contrastive Decoding For Open-ended Text Generation

2022-11-19 · Yixuan Su, Jialu Xu

In the study, we empirically compare the two recently proposed decoding methods, i.e. Contrastive Search (CS) and Contrastive Decoding (CD), for open-ended text generation. The automatic evaluation results suggest that, …

DiversityText Generation

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. Conside…

DiversitySentenceText Generation