Large language models (LLMs) for binary program understanding.
This is all of the code for managing the various datasets, models, and evaluations involved in the Undertale program.
We use Conda for managing dependencies. To set up Undertale with all of its dependencies in a Conda environment, run:
conda env create -f environment.yml
conda activate undertaleFor more details on available scripts and usage information, see the Usage
section of the documentation.
Pull requests and issues are more than welcome. Please review the
Development section of the documentation prior to contributing.
| Date | Venue | Publication |
|---|---|---|
| 04/30/2024 | arXiv | On Training a Neural Network To Explain Binaries |
| 09/15/2025 | IEEE HPEC | Scaling Performance of Large Language Model Pretraining |
DISTRIBUTION STATEMENT A. Approved for public release: distribution unlimited.
© 2023 Massachusetts Institute of Technology
This material is based upon work supported by the Under Secretary of Defense for Research and Engineering under Air Force Contract No. FA8702-15-D-0001. Any opinions, findings, conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the Under Secretary of Defense for Research and Engineering. © 2023 Massachusetts Institute of Technology. The software/firmware is provided to you on an As-Is basis Delivered to the U.S. Government with Unlimited Rights, as defined in DFARS Part 252.227-7013 or 7014 (Feb 2014). Notwithstanding any copyright notice, U.S. Government rights in this work are defined by DFARS 252.227-7013 or DFARS 252.227-7014 as detailed above. Use of this work other than as specifically authorized by the U.S. Government may violate any copyrights that exist in this work.