Share your thoughts, 1 month free Claude Pro on usSee more

Truthfulness on TruthfulQA (accuracy)

72.36Truthfulness Accuracy

AgentRevive

Updated 15d ago

Evaluation Results

Method	Links
AgentRevive 2026.05		72.36
AgentDropout 2026.05		70.15
AgentPrune 2026.05		69.23
G-Designer 2026.05		68.7
ARG-Designer 2026.05		68.44
AutoGen 2026.05		66.63
MASround=T 2026.05		66.05
AgentVerse 2026.05		65.89
AgentRevive 2026.05		65.49
SC (CoT) 2026.05		65.16
CoT 2026.05		64.99
Vanilla 2026.05		64.7
MASround=1 2026.05		64.34
I-DPO + MaPPO 2025.07		63.7
I-DPO + MaPPO 2026.05		63.7
AgentDropout 2026.05		63.62
AgentPrune 2026.05		62.83
G-Designer 2026.05		62.43
ARG-Designer 2026.05		61.78
I-DPO 2025.07		60.7
I-DPO 2026.05		60.7
MASround=T 2026.05		60.14
MASround=1 2026.05		59.88
CoT 2026.05		59.25
DPO + MaPPO 2025.07		59.2
SC (CoT) 2026.05		59.07
DPO + MaPPO 2025.07		58.2
DPO + MaPPO 2026.05		58.2
Vanilla 2026.05		57.59
DPO 2025.07		57.1
IT 2025.07		56.4
I-DPO + MaPPO 2025.07		56.2
I-DPO 2025.07		55.9
Qwen3-4B 2026.04		54.85
IT 2025.07		53.7
IT 2026.05		53.7
Qwen3.5-9B 2026.04		53.69
DPO 2025.07		51.5
DPO 2026.05		51.5
Llama-3.2-3B-Inst. 2026.04		49.78
Qwen3.5-4B 2026.04		48.88
LayerBoost-Qwen3-4B 2026.04		48.03
Gemma-3-4B-it 2026.04		47.85
Jet-Nemotron-4B 2026.04		47.36
Jet-Nemotron-2B 2026.04		46.85
Qwen2.5-1.5B 2026.04		46.58
Qwen3-1.7B 2026.04		45.88
Zamba2-2.7B 2026.04		45.77
Qwen3.5-2B 2026.04		43.42
RWKV7-W3-2.9B 2026.04		43.02
Mamba-2.7B 2026.04		36.03