Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Truthfulness evaluation on TruthfulQA (TruthfulQA Δ)
Loading...
12
TruthfulQA Delta
STM
-0.48
2.76
6
9.24
May 19, 2026
TruthfulQA Delta
Updated 13d ago
Evaluation Results
Method
Method
Links
TruthfulQA Delta
STM
Model=Qwen3-4B
2026.05
12
LoRA
Model=Qwen3-4B
2026.05
10.7
WiSE-FT
Model=Qwen3-4B
2026.05
9.4
L2 Reg
Model=Qwen3-4B
2026.05
8.3
SFT
Model=Qwen3-4B
2026.05
7.8
FLOW
Model=Qwen3-4B
2026.05
7.8
FINCH
Model=Qwen3-4B
2026.05
4
TALR
Model=Qwen3-4B
2026.05
1.6
Base
Model=Qwen3-4B
2026.05
0
DFT
Model=Qwen3-4B
2026.05
0
Feedback
Search any
task
Search any
task