Add MultiScale-Hybrid-e5-both submission

[Vigora]

This PR adds the MultiScale-Hybrid-e5-both detector submission to the RAID leaderboard.

This is a follow-up submission to MultiScale-Hybrid-mpnet (PR #182), using the E5-based variant of the detector. A hybrid detector combining semantic-embedding features with token-probability statistics; method details will be described in a paper currently in preparation (link will be added to metadata.json once available).

The detector was trained on the RAID training split and should be categorized in the "trained on RAID" section.

Submission files:

• metadata.json
• predictions.json

[github-actions]

Eval run succeeded! Link to run: link

Here are the results of the submission(s):

MultiScale-Hybrid-e5-both

Release date: 2026-08-22

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 96.55 and a TPR of 88.85% at FPR=5% and 80.41% at FPR=1%.
Without adversarial attacks, it achieved AUROC of 98.02 and a TPR of 92.83% at FPR=5% and 86.99% 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!