Solexecbench Exploit Blog - #9
Draft
pramodith wants to merge 1 commit into
Draft
Conversation
DeanoC
approved these changes
Jul 28, 2026
DeanoC
left a comment
DeanoC
left a comment
Collaborator
There was a problem hiding this comment.
LGTM
Might be worth making a comment we reported it to NVIDIA as soon as you found it
b-albar
reviewed
Jul 29, 2026
| linkedin: "https://www.linkedin.com/company/112782074/" | ||
| --- | ||
|
|
||
| The gaming of benchmarks in AI is a well-known phenomenon. Sakana AI's "AI CUDA Engineer" was found to have discovered a memory exploit in its evaluation sandbox that let it skip correctness checks entirely, inflating its reported kernel speedups until the company hardened the harness ([Sakana AI](https://x.com/SakanaAILabs/status/1892992938013270019)). Meta faced accusations that Llama 4 Maverick was tuned against benchmark test sets ahead of its release, allegations the company denied ([IT Pro](https://www.itpro.com/technology/artificial-intelligence/meta-llama-4-model-launch-benchmarks)). Test data has also leaked into training sets unintentionally: OpenAI acknowledged a filtering bug that let parts of the GSM8K and MATH test sets end up in GPT-3's training corpus ([Leak, Cheat, Repeat](https://arxiv.org/pdf/2402.03927)). And reward hacking predates LLMs altogether: OpenAI's own CoastRunners agent learned to loop endlessly in a lagoon farming respawning targets instead of finishing the race, racking up a higher score while catching fire and going the wrong way ([OpenAI](https://openai.com/index/faulty-reward-functions/)). In this post, we will discuss a vulnerability that we discovered in the SOLExecbench benchmark that allows submitted kernels to achieve a SOL-score of 1.0, regardless of their actual performance. |
There was a problem hiding this comment.
Competition like the nvfp4 hackaton have also been plagued with this. May be worth to mention to insist on the fact that it is not new but still exists : https://www.gpumode.com/news/reward-hacking-nvfp4
b-albar
approved these changes
Jul 29, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This pull request adds a new blog post detailing a vulnerability in the SOLExecbench benchmark that allows submitted kernels to achieve a perfect SOL-score by exploiting how correctness is checked and trusted. The post provides background on benchmark gaming in AI, a technical breakdown of the vulnerability, proof-of-concept exploit code, and a discussion of the implications for benchmarking robustness.
New blog post: SOLExecbench exploit