CERET: Cost-Effective Extrinsic Refinement for Text Generation

Large Language Models (LLMs) are powerful models for generation tasks, but they may not generate good quality outputs in their first attempt. Apart from model fine-tuning, existing approaches to improve prediction accuracy and quality typically involve LLM self-improvement / self-reflection that inc...

Full description

Saved in:
Bibliographic Details
Published inarXiv.org
Main Authors Cai, Jason, Su, Hang, Sunkara, Monica, Shalyminov, Igor, Saab Mansour
Format Paper Journal Article
LanguageEnglish
Published Ithaca Cornell University Library, arXiv.org 02.11.2024
Subjects
Computational efficiency
Computer Science - Artificial Intelligence
Computer Science - Computation and Language
Computer Science - Learning
Computing costs
Large language models
Online AccessGet full text

Cover

More Information
Summary:Large Language Models (LLMs) are powerful models for generation tasks, but they may not generate good quality outputs in their first attempt. Apart from model fine-tuning, existing approaches to improve prediction accuracy and quality typically involve LLM self-improvement / self-reflection that incorporate feedback from models themselves. Despite their effectiveness, these methods are hindered by their high computational cost and lack of scalability. In this work, we propose CERET, a method for refining text generations by considering semantic stability, entailment and inter-sample uncertainty measures. Experimental results show that CERET outperforms Self-consistency and Self-rerank baselines consistently under various task setups, by ~1.6% in Rouge-1 for abstractive summarization and ~3.5% in hit rate for question answering. Compared to LLM Self-rerank method, our approach only requires 9.4% of its latency and is more cost-effective.
ISSN:2331-8422
DOI:10.48550/arxiv.2406.05588