List your company
ResearchFormerly WizWand

Semantic Textual Similarity on Big Patent (BP)

27.38STS ScorePrompt-Tuning
18.248820.619422.9925.3606May 27, 2026
Updated 5mo ago

Evaluation results

MethodLinks
Prompt-TuningBackbone=Qwen2.5-7B, Evaluation Protocol=Frozen Backbone Methods2026.0527.38
PromptEmbedderBackbone=LLaMA-3.2-3B, Evaluation Protocol=Frozen Backbone Methods2026.0526.92
LoRABackbone=LLaMA-3.2-3B, Evaluation Protocol=Fine-tuning2026.0526.91
Prompt-TuningBackbone=LLaMA-3.2-3B, Evaluation Protocol=Frozen Backbone Methods2026.0526.47
PromptEOLBackbone=Qwen2.5-7B, Evaluation Protocol=Training-free Prompting2026.0526.23
LoRABackbone=Qwen2.5-7B, Evaluation Protocol=Fine-tuning2026.0526.15
PromptEmbedderBackbone=Qwen2.5-7B, Evaluation Protocol=Frozen Backbone Methods2026.0525.9
Pretend-CoTBackbone=Qwen2.5-7B, Evaluation Protocol=Training-free Prompting2026.0525.67
PromptEmbedderBackbone=LLaMA-3.2-1B, Evaluation Protocol=Frozen Backbone Methods2026.0524.85
LoRABackbone=LLaMA-3.2-1B, Evaluation Protocol=Fine-tuning2026.0524.85
Prompt-TuningBackbone=Mistral-7B, Evaluation Protocol=Frozen Backbone Methods2026.0523.91
PromptEOLBackbone=LLaMA-3.2-1B, Evaluation Protocol=Training-free Prompting2026.0523.65
PromptEOLBackbone=LLaMA-3.2-3B, Evaluation Protocol=Training-free Prompting2026.0523.48
EchoBackbone=LLaMA-3.2-3B, Evaluation Protocol=Training-free Prompting2026.0522.38
Pretend-CoTBackbone=LLaMA-3.2-1B, Evaluation Protocol=Training-free Prompting2026.0522.27
EchoBackbone=Qwen2.5-7B, Evaluation Protocol=Training-free Prompting2026.0521.74
PromptEOLBackbone=Mistral-7B, Evaluation Protocol=Training-free Prompting2026.0521.7
Pretend-CoTBackbone=LLaMA-3.2-3B, Evaluation Protocol=Training-free Prompting2026.0521.25
Pretend-CoTBackbone=Mistral-7B, Evaluation Protocol=Training-free Prompting2026.0521.17
EchoBackbone=LLaMA-3.2-1B, Evaluation Protocol=Training-free Prompting2026.0520.74
Prompt-TuningBackbone=LLaMA-3.2-1B, Evaluation Protocol=Frozen Backbone Methods2026.0520.21
EchoBackbone=Mistral-7B, Evaluation Protocol=Training-free Prompting2026.0519.79
LoRABackbone=Mistral-7B, Evaluation Protocol=Fine-tuning2026.0518.83
PromptEmbedderBackbone=Mistral-7B, Evaluation Protocol=Frozen Backbone Methods2026.0518.6