List your company

Faithfulness Evaluation on FaithEval

74.9F1 ScoreProbeRAG
20.71634.78348.8562.917Oct 14, 2025

Evaluation results

MethodLinks
ProbeRAGCategory=Training-Based, Backbone=Qwen3-8B2025.1074.961.6
ProbeRAGCategory=Training-Based, Backbone=Mistral-7B-v0.32025.1074.962.9
ProbeRAGCategory=Training-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1074.464.4
CANOECategory=Training-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1071.656.3
CANOEArchitecture=LLaMA-2-7B-Chat-HF2025.1070.652.3
CANOECategory=Training-Based, Backbone=Qwen3-8B2025.1070.360.2
FaithfulRAGCategory=Prompt-Based, Backbone=Qwen3-8B2025.1070.258.3
CADCategory=Decoding-Based, Backbone=Mistral-7B-v0.32025.1068.933.3
Full-ContextCategory=Baseline, Backbone=Mistral-7B-v0.32025.1068.837.7
ParamMuteCategory=Training-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1068.556.2
AdaCADCategory=Decoding-Based, Backbone=Mistral-7B-v0.32025.1068.434.2
ProbeRAGArchitecture=LLaMA-2-7B-Chat-HF2025.1068.354.4
CANOEArchitecture=Qwen2.5-7B-Instruct2025.1068.153.9
ParamMuteCategory=Training-Based, Backbone=Qwen3-8B2025.1067.858.2
AdaCADCategory=Decoding-Based, Backbone=Qwen3-8B2025.1067.358.2
Context-DPOCategory=Training-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1067.253.7
ParamMuteCategory=Training-Based, Backbone=Mistral-7B-v0.32025.1067.145.2
Full-ContextCategory=Baseline, Backbone=LLaMA-3.1-8B-Instruct2025.1066.953.1
COIECDCategory=Decoding-Based, Backbone=Qwen3-8B2025.1066.656.4
FaithfulRAGCategory=Prompt-Based, Backbone=Mistral-7B-v0.32025.1066.440.3
Context-DPOArchitecture=Qwen2.5-7B-Instruct2025.1065.150.2
Context-DPOCategory=Training-Based, Backbone=Mistral-7B-v0.32025.1064.931.8
KRECategory=Prompt-Based, Backbone=Mistral-7B-v0.32025.1064.836.5
COIECDCategory=Decoding-Based, Backbone=Mistral-7B-v0.32025.1064.429.5
CANOECategory=Training-Based, Backbone=Mistral-7B-v0.32025.1064.144.9
ProbeRAGArchitecture=Qwen2.5-7B-Instruct2025.1063.548.9
Context-DPOArchitecture=LLaMA-2-7B-Chat-HF2025.1063.250.7
FaithfulRAGCategory=Prompt-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1062.449.1
AdaCADCategory=Decoding-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1061.447.7
CADCategory=Decoding-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1059.442.7
KRECategory=Prompt-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1059.112.1
KRECategory=Prompt-Based, Backbone=Qwen3-8B2025.1058.112.3
CADCategory=Decoding-Based, Backbone=Qwen3-8B2025.105728.7
COIECDCategory=Decoding-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1056.141.3
Full-ContextCategory=Baseline, Backbone=Qwen3-8B2025.1055.523.8
Context-DPOCategory=Training-Based, Backbone=Qwen3-8B2025.1055.224
Opin(Instr)Category=Prompt-Based, Backbone=Mistral-7B-v0.32025.1035.714.1
Opin(Instr)Category=Prompt-Based, Backbone=Qwen3-8B2025.103513.9
Opin(Instr)Category=Prompt-Based, Backbone=LLaMA-3.1-8B-Instruct2025.1034.915.1
No-ContextCategory=Baseline, Backbone=LLaMA-3.1-8B-Instruct2025.1027.76
No-ContextCategory=Baseline, Backbone=Mistral-7B-v0.32025.1026.24.4
No-ContextCategory=Baseline, Backbone=Qwen3-8B2025.1022.84.1