Automating Comment Generation for Smart Contract from Bytecode

TL;DR

SmartBT introduces a novel task of generating smart contract comments directly from EVM bytecode, addressing the widespread absence of source code on the Ethereum blockchain. It combines CFG-based structural representations with an IR-augmented semantic retrieval to bridge the gap between low-level bytecode and natural language. The model employs a two-layer encoder–decoder with attention, a copy mechanism for rare terms, and a coverage mechanism to reduce repetition, achieving substantial improvements over baselines and strong human evaluation results. An accompanying dataset of 30,742 <function, bytecode, comment> triplets is released to support future research. Overall, the approach enables meaningful, accurate, and scalable understanding of smart contracts directly from bytecode, with practical implications for end users and developers.

Abstract

Recently, smart contracts have played a vital role in automatic financial and business transactions. To help end users without programming background to better understand the logic of smart contracts, previous studies have proposed models for automatically translating smart contract source code into their corresponding code summaries. However, in practice, only 13% of smart contracts deployed on the Ethereum blockchain are associated with source code. The practical usage of these existing tools is significantly restricted. Considering that bytecode is always necessary when deploying smart contracts, in this paper, we first introduce the task of automatically generating smart contract code summaries from bytecode. We propose a novel approach, named SmartBT (Smart contract Bytecode Translator) for automatically translating smart contract bytecode into fine-grained natural language description directly. Two key challenges are posed for this task: structural code logic hidden in bytecode and the huge semantic gap between bytecode and natural language descriptions. To address the first challenge, we transform bytecode into CFG (Control-Flow Graph) to learn code structural and logic details. Regarding the second challenge, we introduce an information retrieval component to fetch similar comments for filling the semantic gap. Then the structural input and semantic input are used to build an attentional sequence-to-sequence neural network model. The copy mechanism is employed to copy rare words directly from similar comments and the coverage mechanism is employed to eliminate repetitive outputs. The automatic evaluation results show that SmartBT outperforms a set of baselines by a large margin, and the human evaluation results show the effectiveness and potential of SmartBT in producing meaningful and accurate comments for smart contract code from bytecode directly.

Figures (8)

• Figure 1: Life Cycle of Smart Contracts
• Figure 2: Motivating Example of Using SmartBT
• Figure 3: Workflow of SmartBT
• Figure 4: Example of Solidity Source Code, Bytecode and CFG
• Figure 5: Length Distribution of The Training Data