NUWA DEVELOPER DISCLOSURE · Mistral AI
Shieldstral
Developer documentation covering Shieldstral.
Published Not verified
Sources ↓Developer-authored disclosure. Selected source sections were read; numerical results and comparability have not been audited. NUWA has not independently replicated these experiments.
Models covered
behavior · Original source → Safety classification, refusal detection and multimodal safety evaluations
Protocol record →HarmBench
Attack-dependent generation of harmful responses.
A protocol mention does not establish cross-report comparability or independent replication.
This section is catalogued; no precise numerical result has been extracted here. Open the original report for its figures and full findings.
Report coverage
| Section | Source location | Treatment |
|---|---|---|
| Safety classification, refusal detection and multimodal safety evaluations | Original source → Safety classification, refusal detection and multimodal safety evaluations | Linked protocol Policy-adaptive safety classifier; its moderation results do not establish the safety of Mistral Large 4. |
References
- [1] Shieldstral