johannes-garstenauer/

distilbert_masking_heaps

DistilBERT for masked language modelling trained on OpenSSH heap data structures dataset for the purpose of generating representations. This model was created for the thesis "Generating Robust Representations of Structures in OpenSSH Heap Dumps" by Johannes Garstenauer.

Training data: https://huggingface.co/datasets/johannes-garstenauer/structs_token_size_4_reduced_labelled_train Validation data: https://huggingface.co/datasets/johannes-garstenauer/structs_token_size_4_reduced_labelled_eval