Skip to content

Submission/v24 3k#138

Open
MohamedMady19 wants to merge 5 commits into
liamdugan:mainfrom
MohamedMady19:submission/v24-3k
Open

Submission/v24 3k#138
MohamedMady19 wants to merge 5 commits into
liamdugan:mainfrom
MohamedMady19:submission/v24-3k

Conversation

@MohamedMady19
Copy link
Copy Markdown

No description provided.

@github-actions
Copy link
Copy Markdown

ghost commented May 12, 2026

Eval run succeeded! Link to run: link

Here are the results of the submission(s):

DeBERTa-ConPara-v2.2-Seed42

Release date: 2026-05-01

I've committed detailed results of this detector's performance on the test set to this PR.

On the RAID dataset as a whole (aggregated across all generation models, domains, decoding strategies, repetition penalties, and adversarial attacks), it achieved an AUROC of 96.86 and a TPR of 97.29% at FPR=5% and 93.55% at FPR=1%.
Without adversarial attacks, it achieved AUROC of 97.05 and a TPR of 97.02% at FPR=5% and 93.39% at FPR=1%.

DeBERTa-ConPara-v2.4-3000cell

Release date: 2026-05-11

I've committed detailed results of this detector's performance on the test set to this PR.

On the RAID dataset as a whole (aggregated across all generation models, domains, decoding strategies, repetition penalties, and adversarial attacks), it achieved an AUROC of 97.35 and a TPR of 97.79% at FPR=5% and 94.95% at FPR=1%.
Without adversarial attacks, it achieved AUROC of 97.26 and a TPR of 97.59% at FPR=5% and 94.64% at FPR=1%.

If all looks well, a maintainer will come by soon to merge this PR and your entry/entries will appear on the leaderboard. If you need to make any changes, feel free to push new commits to this PR. Thanks for submitting to RAID!

@MohamedMady19
Copy link
Copy Markdown
Author

Eval run succeeded! Link to run: link

Here are the results of the submission(s):

DeBERTa-ConPara-v2.2-Seed42

Release date: 2026-05-01

I've committed detailed results of this detector's performance on the test set to this PR.

On the RAID dataset as a whole (aggregated across all generation models, domains, decoding strategies, repetition penalties, and adversarial attacks), it achieved an AUROC of 96.86 and a TPR of 97.29% at FPR=5% and 93.55% at FPR=1%. Without adversarial attacks, it achieved AUROC of 97.05 and a TPR of 97.02% at FPR=5% and 93.39% at FPR=1%.

DeBERTa-ConPara-v2.4-3000cell

Release date: 2026-05-11

I've committed detailed results of this detector's performance on the test set to this PR.

On the RAID dataset as a whole (aggregated across all generation models, domains, decoding strategies, repetition penalties, and adversarial attacks), it achieved an AUROC of 97.35 and a TPR of 97.79% at FPR=5% and 94.95% at FPR=1%. Without adversarial attacks, it achieved AUROC of 97.26 and a TPR of 97.59% at FPR=5% and 94.64% at FPR=1%.

If all looks well, a maintainer will come by soon to merge this PR and your entry/entries will appear on the leaderboard. If you need to make any changes, feel free to push new commits to this PR. Thanks for submitting to RAID!

Thank you, Please feel free to merge it

@liamdugan
Copy link
Copy Markdown
Owner

liamdugan commented May 12, 2026

Hi @MohamedMady19 so currently you have five submissions to the leaderboard DeBERTa-ConPara through DeBERTa-ConPara-v3. These two submissions would give you 7 on the leaderboard.

If any of these submissions are outdated, can you please delete them as part of your PR? the submission folders should be in leaderboard/submissions

@MohamedMady19
Copy link
Copy Markdown
Author

Hi @MohamedMady19 so currently you have five submissions to the leaderboard DeBERTa-ConPara through DeBERTa-ConPara-v3. These two submissions would give you 7 on the leaderboard.

If any of these submissions are outdated, can you please delete them as part of your PR? the submission folders should be in leaderboard/submissions

Hi @liamdugan, I've cleaned up the outdated submissions and kept only the four that are relevant to our current paper:

DeBERTa-V22-Seed42 - main paper result (97.29% TPR@5%)
DeBERTa-V23-Novel - GDWL training objective (97.69%)
DeBERTa-V24-3000cell - RAID scaling study (97.79%)
DeBERTa-V24-5000cell - RAID scaling study (pending evaluation)

The earlier submissions (v1 through v3, Con-L01) have been removed. Thanks!

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants