Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Poisoning Defense on RAG Evaluation Datasets NQ, PubMedQA, TriviaQA
Loading...
59.4
Contextual Recall
TrustRAG
25.912
34.606
43.3
51.994
Apr 22, 2026
Contextual Recall
ASR
Faithfulness
Updated 1mo ago
Evaluation Results
Method
Method
Links
Contextual Recall
ASR
Faithfulness
TrustRAG
Mode=Static Targeted
2026.04
59.4
0
79
ADO
Controller Model=Mistral
2026.04
47
4
77
ADO
Controller Model=Qwen 3
2026.04
43.9
44
73
ADO
Controller Model=Llama 3
2026.04
42.9
0
75.3
Full Stack
Mode=Static
2026.04
34.3
-
74.5
ADO
Controller Model=Gemma 3
2026.04
29.7
1
76
ADO
Controller Model=GPT-4o
2026.04
27.2
35
77
Feedback
Search any
task
Search any
task
