English

CERET: Cost-Effective Extrinsic Refinement for Text Generation

Computation and Language 2024-11-05 v2 Artificial Intelligence Machine Learning

Abstract

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.

Keywords

Cite

@article{arxiv.2406.05588,
  title  = {CERET: Cost-Effective Extrinsic Refinement for Text Generation},
  author = {Jason Cai and Hang Su and Monica Sunkara and Igor Shalyminov and Saab Mansour},
  journal= {arXiv preprint arXiv:2406.05588},
  year   = {2024}
}

Comments

The source code and data samples are released at https://github.com/amazon-science/CERET-LLM-refine

R2 v1 2026-06-28T16:58:25.491Z