Question Answering on TQA (Acc., ASR)
76.8AccuracyReliabilityRAG
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ReliabilityRAGTop-k=50, Pos=50, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 76.8 | 5.6 | |
| ReliabilityRAGTop-k=50, Pos=25, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 75.8 | 9 | |
| ReliabilityRAGTop-k=50, Pos=Pos 25, Generator LLM=Grok-4-fast2026.05 | 74.6 | 8.8 | |
| RADARTop-k=50, Pos=50, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 72.2 | 16 | |
| ReliabilityRAGTop-k=50, Pos=Pos 50, Generator LLM=Grok-4-fast2026.05 | 71.8 | 4.2 | |
| RADARTop-k=50, Pos=Pos 50, Generator LLM=Grok-4-fast2026.05 | 71.8 | 10 | |
| RADARTop-k=50, Pos=25, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 71.2 | 16.2 | |
| RobustRAGTop-k=50, Pos=50, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 71.2 | 16.8 | |
| RobustRAGTop-k=50, Pos=25, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 71 | 17 | |
| RADARTop-k=10, Pos=Pos 10, Generator LLM=Grok-4-fast2026.05 | 71 | 12.8 | |
| AstuteRAGTop-k=10, Pos=Pos 10, Generator LLM=Grok-4-fast2026.05 | 70.8 | 3.6 | |
| ReliabilityRAGTop-k=10, Pos=Pos 10, Generator LLM=Grok-4-fast2026.05 | 70.2 | 15.2 | |
| RADARTop-k=50, Pos=Pos 25, Generator LLM=Grok-4-fast2026.05 | 70.2 | 10 | |
| AstuteRAGTop-k=50, Pos=25, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 70 | 16.4 | |
| RobustRAGTop-k=50, Pos=Pos 25, Generator LLM=Grok-4-fast2026.05 | 70 | 10.2 | |
| RobustRAGTop-k=50, Pos=Pos 50, Generator LLM=Grok-4-fast2026.05 | 69.8 | 5.6 | |
| RADARTop-k=10, Pos=10, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 69.4 | 24.2 | |
| RobustRAGTop-k=10, Pos=10, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 69 | 18.4 | |
| RobustRAGTop-k=10, Pos=Pos 10, Generator LLM=Grok-4-fast2026.05 | 69 | 18.4 | |
| ReliabilityRAGTop-k=10, Pos=10, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 68.8 | 22.4 | |
| AstuteRAGTop-k=50, Pos=50, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 68.8 | 11.2 | |
| RADARTop-k=50, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 67.6 | 13.8 | |
| InstructRAGTop-k=50, Pos=Pos 50, Generator LLM=Grok-4-fast2026.05 | 67.6 | 4.4 | |
| RobustRAGTop-k=50, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 67.2 | 15.8 | |
| AstuteRAGTop-k=50, Pos=Pos 25, Generator LLM=Grok-4-fast2026.05 | 66.8 | 3.2 | |
| AstuteRAGTop-k=50, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 66.6 | 3.8 | |
| AstuteRAGTop-k=50, Pos=Pos 50, Generator LLM=Grok-4-fast2026.05 | 66.6 | 3 | |
| RobustRAGTop-k=10, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 66 | 16.2 | |
| RADARTop-k=10, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 66 | 18.8 | |
| InstructRAGTop-k=50, Pos=Pos 25, Generator LLM=Grok-4-fast2026.05 | 66 | 4.4 | |
| AstuteRAGTop-k=10, Pos=10, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 65.2 | 13 | |
| AstuteRAGTop-k=10, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 65 | 3.8 | |
| InstructRAGTop-k=10, Pos=Pos 10, Generator LLM=Grok-4-fast2026.05 | 64.4 | 5.2 | |
| RADARTop-k=50, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 63.6 | 32 | |
| RobustRAGTop-k=50, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 63.5 | 28.2 | |
| ReliabilityRAGTop-k=10, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 62.6 | 30.6 | |
| RADARTop-k=10, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 61.4 | 33.4 | |
| RobustRAGTop-k=10, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 61.2 | 23.4 | |
| InstructRAGTop-k=10, Pos=10, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 60.8 | 28.6 | |
| ReliabilityRAGTop-k=50, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 60.8 | 35.2 | |
| AstuteRAGTop-k=50, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 59.4 | 30.4 | |
| ReliabilityRAGTop-k=50, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 59.4 | 37.6 | |
| InstructRAGTop-k=50, Pos=25, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 58.2 | 17.1 | |
| ReliabilityRAGTop-k=10, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 58 | 40.6 | |
| InstructRAGTop-k=50, Pos=50, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 57.2 | 35 | |
| AstuteRAGTop-k=10, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 55.2 | 35.2 | |
| InstructRAGTop-k=10, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 51.6 | 25.2 | |
| InstructRAGTop-k=50, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 49.8 | 30.4 | |
| InstructRAGTop-k=10, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 48.8 | 41.4 | |
| InstructRAGTop-k=50, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 48.4 | 38.4 | |
| Vanilla RAGTop-k=50, Pos=25, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 47.2 | 37.8 | |
| Vanilla RAGTop-k=50, Pos=50, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 46.2 | 32.8 | |
| Vanilla RAGTop-k=50, Pos=Pos 50, Generator LLM=Grok-4-fast2026.05 | 43.2 | 43.6 | |
| Vanilla RAGTop-k=50, Pos=Pos 25, Generator LLM=Grok-4-fast2026.05 | 41 | 55.6 | |
| Vanilla RAGTop-k=10, Pos=10, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 38.8 | 43.4 | |
| Vanilla RAGTop-k=10, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 36.8 | 50.6 | |
| Vanilla RAGTop-k=10, Pos=Pos 10, Generator LLM=Grok-4-fast2026.05 | 33.4 | 63.8 | |
| Vanilla RAGTop-k=50, Pos=1, Attack Type=PIA attack, Generator LLM=GPT-4o2026.05 | 32.6 | 50.6 | |
| Vanilla RAGTop-k=10, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 11 | 89 | |
| Vanilla RAGTop-k=50, Pos=Pos 1, Generator LLM=Grok-4-fast2026.05 | 8 | 92 |