afvanluenen's picture
Update README.md
1b02d6b verified
|
Raw
History Blame
1.98 kB
metadata
license: eupl-1.2
tags:
  - assembly
  - arm64
  - amd64
  - risc-v
  - i386

Model Description

ASMTransformers is a project to train and use a machine learning model to compare assembly (ARM64, AMD64, RISC-V, i386) functions to a database of known functions, to aid in the process of reverse engineering.

Status

?

Relevant links

Version

TODO: calver of new release

Usage

TODO: do we refer to the inference file of our repo?

Intended use

The model has been trained and tested to be used for similarity search of assembly code. It has not been trained/tested on any other languages than ARM64, AMD64, RISC-V or I386, nor has it been tested on other downstream tasks.

Architecture description

TODO: get model architecture like on the model card of the previous model Don't forget to describe that during pretraining, the --mlm-prob parameter has been set to 0.4, in accordance to [paper](INSERT LINK) Parameters used: epochs=19, eval_steps=10000, batch_size=512, gradient_accumulation_steps=1, mlm_prob=0.4, bf16=True, tf32=True

add estimated time it has cost to train: pretraining: finetuning: 4 hours on 1 NVIDIA H200

Output

The model outputs embeddings that can be compared using cosine similarity

Data

TODO: how did we get the data?

Preprocessing

Several preprocessing steps have been taken, that differ slightly between architectures. TODO: update to same as architecture.md in repo

Performance

Performance was measured in two ways: Mean Reciprocal Rank (MRR) and Accuracy@1


|model|ARM64 mrr|ARM64 acc|AMD64 mrr|AMD64 acc|RISC-V mrr|RISC-V acc|I386 mrr|I386 acc|crosslingual mrr|crosslingual acc|