Indirect Prompt Injection on LivePI Total
10.7ASRClaude Opus 4.6
Evaluation Results
| Method | Links | |
|---|---|---|
| Claude Opus 4.6Model Type=Base LLM2026.05 | 10.7 | |
| Kimi K2.5Model Type=Base LLM2026.05 | 16.6 | |
| GPT-5.3-CodexModel Type=Base LLM2026.05 | 27.2 | |
| GLM-5Model Type=Base LLM2026.05 | 27.8 | |
| Gemini 3.1 ProModel Type=Base LLM2026.05 | 29.6 |