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

Information Retrieval Reasoning on NevIR

0.617Pairwise Accuracy

GPT-4o

Updated 3mo ago

Evaluation Results

Method	Links
GPT-4o 2026.04		0.617
OpenAI o1 2026.04		0.597
Llama 3.1 70B 2026.04		0.588
MonoT5 3B 2026.04		0.506
Llama 3.1 8B 2026.04		0.422
Qwen2 7B 2026.04		0.398
Llama 3.1 8B 2026.04		0.379
RoBERTa-large 2026.04		0.343
Qwen2 7B 2026.04		0.338
Llama 3.1 8B 2026.04		0.306
Qwen2 7B 2026.04		0.298
RoBERTa-large 2026.04		0.291
stsb-roberta-large 2026.04		0.249
RoBERTa-large 2026.04		0.245