List your company

Evidence Extraction on UCLH dataset (test)

77.17R-L Scorecopy
48.3155.802563.29570.7875Jun 11, 2026

Evaluation results

MethodLinks
copyModel=Qwen2.5-32B-Instruct, AnsLen=39.752026.0677.1781.9980.5477.6282.382.7578.71
copyModel=GPT-OSS-120B, AnsLen=44.962026.0675.3778.8482.7676.6779.7984.8578.07
safetyModel=GPT-OSS-120B, AnsLen=45.052026.0670.8273.7180.1471.6974.2481.7472.95
linesModel=Qwen2.5-32B-Instruct, AnsLen=93.912026.066863.4990.468.265.8491.2870.36
safetyModel=Qwen2.5-32B-Instruct, AnsLen=39.442026.0667.9471.6375.568.9173.4176.7670.98
linesModel=GPT-OSS-120B, AnsLen=109.492026.0662.3153.894.0662.5356.3194.8565.4
juniorModel=GPT-OSS-120B, AnsLen=118.002026.0649.8342.0587.4550.8544.1487.5352.98
juniorModel=Qwen2.5-32B-Instruct, AnsLen=146.332026.0649.4244.1583.1750.5246.9283.5453.35