LinguaGuard — original wording balanced accuracy — TS-Guard native XML step prompt: leaderboard
Original-wording TSBench AgentDojo: 1,220 steps, 868 safe, 352 unsafe, 589 trajectories, 58 distinct user instructions. Balanced accuracy of supplied-action risk classification. Condition: TS-Guard native XML step prompt. Managed models three repeats; self-hosted greedy single run. Appendix E explicitly identifies Qwen3.8-27B FP8 and Sonnet5 thinking disabled. Native trajectory guards predict prefix-level risk against step labels and are not pooled with step-level guards. Scanner withholds user request. Provider defaults do not resolve unknown checkpoint variants; qualified names remain distinct. FPR/TPR and translation shifts are operating-point diagnostics, not standalone general-capability scores. Metric: Balanced accuracy. Higher is better. Source: https://arxiv.org/html/2610.22244v1
Metric: Balanced accuracy. Source: arxiv.org. 1 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | TS-Guard 7.6B | 0.91 |
Interactive version: theaggregate.ai/benchmark?slug=linguaguard-original-wording-balanced-accuracy-ts-guard-native-xml-step-prompt · How It Works · Data refreshed daily, snapshot 2026-10-17.