Relation Extraction on MTSamples (test) benchmark leaderboard · SOTA2 Research

Relation Extraction on MTSamples (test)

83.1Strict F1 ScoreCentralized fine-tuning

Evaluation Results

Method
Centralized fine-tuning2026.01	83.1	91.3
Centralized fine-tuning2026.01	82.9	91.1
Fed-MedLoRA+2026.01	82.6	85.6
Fed-MedLoRA2026.01	81.7	84.9
LLaMA3-8B Centralized2026.01	78.8	85.5
DeepSeek-R1-Distill-8B Centralized2026.01	78.5	85.5
Fed-MedLoRA+2026.01	77.3	90.7
Fed-MedLoRA2026.01	76.7	89.2
Fed-MedLoRA+2026.01	76.2	81.6
Fed-MedLoRA2026.01	75.9	79.7
Fed-MedLoRA2026.01	73.9	82.1
Fed-MedLoRA+2026.01	73.6	89.6
DeepSeek-R1-Distill-8B Single site2026.01	72.5	79.9
Single site fine-tuning2026.01	72.1	79.8
Single site fine-tuning2026.01	71.9	79.3
LLaMA3-8B Single site2026.01	70.3	78
Centralized fine-tuning2026.01	58.4	65.7
Bio_ClinicalBERT Centralized2026.01	51.5	62.1
Single site fine-tuning2026.01	47.4	54.9
Bio_ClinicalBERT Single site2026.01	43.3	50
Zero-shot2026.01	10.6	13.9
DeepSeek-R1-Distill-8B Zero-shot2026.01	10.6	13.9
Zero-shot2026.01	9.6	17.8
GPT-4o Zero-shot2026.01	9.6	17.8
Zero-shot2026.01	5.6	7.4
LLaMA3-8B Zero-shot2026.01	5.6	7.4