gennm-ndss-ae

GenNm is a tool that employs a large language model-based technique for recovering meaningful variable names from stripped binary code, which typically lacks semantic identifiers. Its primary use case is to enhance the reverse engineering process, making decompiled code more understandable and maintainable. Notable features include detailed artifact evaluation scripts, the ability to run training and inference with specific GPU requirements, and comprehensive installation instructions to facilitate reproducibility in experimental results.