This website requires JavaScript.

Automated Smart Contract Summarization via LLMs

Yingjie MaoXiaoqi LiZongwei LiWenkai Li
Feb 2024
0被引用
0笔记
摘要原文
Automatic code Summarization generation technology is widely used in the development and maintenance of smart contracts. In recent years, with the advent of Large Language Models (LLMs), Gemini has received a lot of attention as the first Large Multimodal models (LMMs) to support multimodal input. However, it is unclear how LMMs can generate contract code summarization from multimodal inputs. In this paper, we focus on evaluating Gemini on real-world smart contracts, comparing it to the MMTrans, and exploring how to combine multimodal prompts to generate a contract code summarization. We used several widely used metrics (BLEU, METEOR, and ROUGE-L) to measure the quality of the generated summarization. Our experiments show that METEOR and ROUGEL metrics, Gemini-Pro-Vision achieves 21.17% and 21.05% scores for code comments generated by three-shot prompts. These scores are better than those generated by one-shot and five-shot prompts.
展开全部
机器翻译
AI理解论文&经典十问
图表提取
参考文献
发布时间 · 被引用数 · 默认排序
被引用
发布时间 · 被引用数 · 默认排序
社区问答