Multi-modal Learning for WebAssembly Reverse Engineering

The increasing adoption of WebAssembly (Wasm) for performancecritical and security-sensitive tasks drives the demand for Web-Assembly program comprehension and reverse engineering.Recent studies have introduced machine learning (ML)-based Web-Assembly reverse engineering tools.Yet, the generalization of taskspecific ML solutions remains challenging, because their effectiveness hinges on the availability of an ample supply of high-quality task-specific labeled data.Moreover, previous works trained models only with features extracted from WebAssembly, overlooking the high-level semantics present in the corresponding source code and its documentation.Acknowledging the abundance of available source code with documentation, which can be compiled into Web-Assembly, we propose to learn representations of them concurrently and harness their mutual relationships for effective WebAssembly reverse engineering.In this paper, we present WasmRev, the first multi-modal pretrained language model for WebAssembly reverse engineering.WasmRev is pre-trained using self-supervised learning on a largescale multi-modal corpus encompassing source code, code documentation and the compiled WebAssembly, without requiring labeled data.WasmRev incorporates three tailored multi-modal pretraining tasks to capture various characteristics of WebAssembly and cross-modal relationships.WasmRev is only trained once to produce general-purpose representations that can broadly support WebAssembly reverse engineering tasks through few-shot finetuning with much less labeled data, improving data efficiency.We fine-tune WasmRev onto three important reverse engineering tasks: type recovery, function purpose identification and WebAssembly summarization.Our results show that WasmRev pre-trained on the corpus of multi-modal samples establishes a robust foundation for these tasks, achieving high task accuracy and outperforming the state-of-the-art ML methods for WebAssembly reverse engineering.