List your company
ResearchFormerly WizWand

Text Classification on SST2 (ACC)

97.36AccuracyLS-unLLaMA-2-7B
52.577664.203875.8387.4562Oct 2, 2023Feb 26, 2024Jul 22, 2024Dec 16, 2024May 12, 2025Oct 6, 2025Mar 2, 2026
Updated 8mo ago

Evaluation results

MethodLinks
LS-unLLaMA-2-7BModel Size=7B2023.1097.36
PlaintextBase Model=Qwen2.5-14B-Instruct2026.0397.25
AloePriBase Model=Qwen2.5-14B-Instruct2026.0397.13
LS-LLaMA-2-13BModel Size=13B2023.1096.9
LS-LLaMA-2-7BModel Size=7B2023.1096.67
RoBERTa-LargeModel Architecture=Discriminant baseline2023.1096.1
SCModel=Llama, k-shot=8, Seeds=52025.0595.39
SCModel=Mistral, Shots=42025.0595.23
CCModel=Mistral, Shots=42025.0594.92
DCModel=Mistral, k-shot=8, Seeds=52025.0594.84
CCModel=Llama, k-shot=8, Seeds=52025.0594.77
RoBERTa-BaseModel Architecture=Discriminant baseline2023.1094.61
DCModel=Llama, k-shot=8, Seeds=52025.0594.61
Base LLMModel=Mistral, k-shot=8, Seeds=52025.0594.53
SCModel=Llama, Shots=42025.0594.45
CCModel=Mistral, k-shot=8, Seeds=52025.0594.38
DCModel=Mistral, Shots=42025.0594.38
Base LLMModel=Llama, k-shot=8, Seeds=52025.0594.3
BCModel=Mistral, k-shot=8, Seeds=52025.0594.3
BCModel=Llama, k-shot=8, Seeds=52025.0594.22
Base LLMModel=Mistral, Shots=42025.0594.22
BCModel=Mistral, Shots=42025.0594.14
DCModel=Llama, Shots=42025.0593.83
BCModel=Llama, Shots=42025.0593.75
Base LLMModel=Llama, Shots=42025.0593.63
DP-ForwardBase Model=Qwen2.5-14B-Instruct2026.0393.6
SCModel=Mistral, k-shot=8, Seeds=52025.0593.59
EspeWAttack=No Attack2025.1293.54
WARDENAttack=Detect2025.1293.5
CCModel=Llama, Shots=42025.0593.48
EspeWAttack=Detect2025.1293.43
SemMarkAttack=Dim2025.1293.43
GPT-3 175BEvaluation Protocol=few-shot, Model Size=175B2023.1093.4
OriginalAttack=None2025.1293.39
EmbMarkerAttack=No Attack2025.1293.35
WETAttack=Dim2025.1293.35
SemMarkAttack=No Attack2025.1293.31
SemMarkAttack=Detect2025.1293.27
AGNModel Architecture=Discriminant baseline2023.1093.27
WARDENAttack=No Attack2025.1293.2
WETAttack=Detect2025.1293.2
EmbMarkerAttack=Dim2025.1293.12
WETAttack=No Attack2025.1293.08
EmbMarkerAttack=Detect2025.1293
BERT-LargeModel Architecture=Discriminant baseline2023.1092.86
BERT-BaseModel Architecture=Discriminant baseline2023.1092.78
WARDENAttack=Dim2025.1292.77
LS-unLLaMA-2-13BModel Size=13B2023.1092.77
SGTBase Model=Qwen2.5-14B-Instruct2026.0392.43
EspeWAttack=Dim2025.1292.35
CCModel=Qwen, Shots=42025.0591.99
LLaMA-2-7BEvaluation Protocol=instruction-tuning, Model Size=7B2023.1091.97
SCModel=Qwen, k-shot=8, Seeds=52025.0591.88
SCModel=Qwen, Shots=42025.0590.82
EmbMarkerAttack=CSE2025.1290.48
CCModel=Qwen, k-shot=8, Seeds=52025.0589.53
SemMarkAttack=CSE2025.1289.41
BCModel=Qwen, Shots=42025.0589.06
WARDENAttack=CSE2025.1288.34
Base LLMModel=Qwen, Shots=42025.0587.93
RANTEXTBase Model=Qwen2.5-14B-Instruct2026.0387.5
DCModel=Qwen, Shots=42025.0587.3
EspeWAttack=CSE2025.1286.81
WETAttack=CSE2025.1286.81
BCModel=Qwen, k-shot=8, Seeds=52025.0586.52
SANTEXTBase Model=Qwen2.5-14B-Instruct2026.0381.31
LLaMA-2-7BEvaluation Protocol=zero-shot, Model Size=7B2023.1076.26
Base LLMModel=Qwen, k-shot=8, Seeds=52025.0576.09
DCModel=Qwen, k-shot=8, Seeds=52025.0571.72
LLaMA-2-13BEvaluation Protocol=zero-shot, Model Size=13B2023.1069.9
GPT-3 175BEvaluation Protocol=zero-shot, Model Size=175B2023.1054.3