List your company

Truthfulness on TruthfulQA (accuracy)

72.36Truthfulness AccuracyAgentRevive
16.761631.195845.6360.0642Jul 27, 2025Sep 21, 2025Nov 16, 2025Jan 11, 2026Mar 8, 2026May 3, 2026Jun 28, 2026

Evaluation results

MethodLinks
AgentReviveBase model=Deepseek-V3-671B-Instruct, Var. NS=true, Flex. State=true2026.0572.36
AgentDropoutBase model=Deepseek-V3-671B-Instruct, Var. NS=true, Flex. State=false2026.0570.15
AgentPruneBase model=Deepseek-V3-671B-Instruct, Var. NS=false, Flex. State=false2026.0569.23
G-DesignerBase model=Deepseek-V3-671B-Instruct, Var. NS=false, Flex. State=false2026.0568.7
ARG-DesignerBase model=Deepseek-V3-671B-Instruct, Var. NS=true, Flex. State=false2026.0568.44
AutoGenBase model=Deepseek-V3-671B-Instruct, Var. NS=false, Flex. State=false2026.0566.63
MASround=TBase model=Deepseek-V3-671B-Instruct, Var. NS=false, Flex. State=false2026.0566.05
AgentVerseBase model=Deepseek-V3-671B-Instruct, Var. NS=false, Flex. State=false2026.0565.89
AgentReviveBase model=Llama3-8B-Instruct, Var. NS=true, Flex. State=true2026.0565.49
SC (CoT)Base model=Deepseek-V3-671B-Instruct, Var. NS=false, Flex. State=false2026.0565.16
CoTBase model=Deepseek-V3-671B-Instruct, Var. NS=false, Flex. State=false2026.0564.99
VanillaBase model=Deepseek-V3-671B-Instruct, Var. NS=false, Flex. State=false2026.0564.7
MASround=1Base model=Deepseek-V3-671B-Instruct, Var. NS=false, Flex. State=false2026.0564.34
I-DPO + MaPPOModel=Llama-3-8B-Instruct2025.0763.7
I-DPO + MaPPOBase Model=Llama-3-8B-Instruct, DPO Mode=iterative2026.0563.7
AgentDropoutBase model=Llama3-8B-Instruct, Var. NS=true, Flex. State=false2026.0563.62
AgentPruneBase model=Llama3-8B-Instruct, Var. NS=false, Flex. State=false2026.0562.83
G-DesignerBase model=Llama3-8B-Instruct, Var. NS=false, Flex. State=false2026.0562.43
ARG-DesignerBase model=Llama3-8B-Instruct, Var. NS=true, Flex. State=false2026.0561.78
I-DPOModel=Llama-3-8B-Instruct2025.0760.7
I-DPOBase Model=Llama-3-8B-Instruct, DPO Mode=iterative2026.0560.7
MASround=TBase model=Llama3-8B-Instruct, Var. NS=false, Flex. State=false2026.0560.14
MASround=1Base model=Llama3-8B-Instruct, Var. NS=false, Flex. State=false2026.0559.88
CoTBase model=Llama3-8B-Instruct, Var. NS=false, Flex. State=false2026.0559.25
DPO + MaPPOModel=Qwen2.5-7B-Instruct2025.0759.2
SC (CoT)Base model=Llama3-8B-Instruct, Var. NS=false, Flex. State=false2026.0559.07
DPO + MaPPOModel=Llama-3-8B-Instruct2025.0758.2
DPO + MaPPOBase Model=Llama-3-8B-Instruct, DPO Mode=offline2026.0558.2
VanillaBase model=Llama3-8B-Instruct, Var. NS=false, Flex. State=false2026.0557.59
DPOModel=Qwen2.5-7B-Instruct2025.0757.1
ITModel=Qwen2.5-7B-Instruct2025.0756.4
I-DPO + MaPPOModel=Qwen2.5-7B-Instruct2025.0756.2
I-DPOModel=Qwen2.5-7B-Instruct2025.0755.9
Qwen3-4BArchitecture Type=O(n^2)2026.0454.85
ITModel=Llama-3-8B-Instruct2025.0753.7
ITBase Model=Llama-3-8B-Instruct2026.0553.7
Qwen3.5-9BArchitecture Type=Hybrid2026.0453.69
QCBModel=Phi-2-2.7B, Quantization=INT82026.0652.5
QCBModel=Phi-2-2.7B, Quantization=NF42026.0652.5
QCBModel=Phi-2-2.7B, Quantization=FP42026.0651.7
QCBModel=LLaMA3-8B, Quantization=INT82026.0651.7
DPOModel=Llama-3-8B-Instruct2025.0751.5
DPOBase Model=Llama-3-8B-Instruct, DPO Mode=offline2026.0551.5
Llama-3.2-3B-Inst.Architecture Type=O(n^2)2026.0449.78
QCBModel=LLaMA3-8B, Quantization=NF42026.0649.1
Qwen3.5-4BArchitecture Type=Hybrid2026.0448.88
LayerBoost-Qwen3-4BArchitecture Type=Hybrid2026.0448.03
Gemma-3-4B-itArchitecture Type=Hybrid2026.0447.85
Jet-Nemotron-4BArchitecture Type=Hybrid2026.0447.36
QuantGuardModel=LLaMA3-8B, Quantization=INT82026.0647.1
QuantGuardModel=Phi-2-2.7B, Quantization=INT82026.0646.9
Jet-Nemotron-2BArchitecture Type=Hybrid2026.0446.85
QCBModel=LLaMA3-8B, Quantization=FP42026.0646.6
Qwen2.5-1.5BArchitecture Type=O(n^2)2026.0446.58
QuantGuardModel=Phi-2-2.7B, Quantization=FP42026.0646.1
Qwen3-1.7BArchitecture Type=O(n^2)2026.0445.88
Zamba2-2.7BArchitecture Type=Hybrid2026.0445.77
QuantGuardModel=Phi-2-2.7B, Quantization=NF42026.0645.6
QuantGuardModel=LLaMA3-8B, Quantization=NF42026.0644.6
QuantGuardModel=LLaMA3-8B, Quantization=FP42026.0643.5
Qwen3.5-2BArchitecture Type=Hybrid2026.0443.42
CleanModel=LLaMA3-8B, Quantization=INT82026.0643.2
RWKV7-W3-2.9BArchitecture Type=O(n)2026.0443.02
CleanModel=Phi-2-2.7B, Quantization=NF42026.0641.5
CleanModel=Phi-2-2.7B, Quantization=INT82026.0641.3
CleanModel=Phi-2-2.7B, Quantization=FP42026.0639.9
CleanModel=LLaMA3-8B, Quantization=FP42026.0639.9
CleanModel=LLaMA3-8B, Quantization=NF42026.0639.8
Mamba-2.7BArchitecture Type=O(n)2026.0436.03
QuantGuardModel=Gemma-2B, Quantization=INT82026.0621.4
QCBModel=Gemma-2B, Quantization=FP42026.0621.3
CleanModel=Gemma-2B, Quantization=INT82026.0620.4
QuantGuardModel=Gemma-2B, Quantization=NF42026.0620.3
CleanModel=Gemma-2B, Quantization=NF42026.0620.1
QuantGuardModel=Gemma-2B, Quantization=FP42026.0620
CleanModel=Gemma-2B, Quantization=FP42026.0619.5
QCBModel=Gemma-2B, Quantization=NF42026.0619.5
QCBModel=Gemma-2B, Quantization=INT82026.0618.9