ResearchBenchmarksMulti-task Evaluation on Average (test)Follow52.45Hit ScoreNA-SFT36.933240.961644.9949.0184May 28, 2024Evaluation ResultsMethodMethodLinksHit ScoreFalse Alarm RateNA-SFTBackbone=Mistral-7BBackbone=Mistral-7B2024.0552.4568.65SFTBackbone=Mistral-7BBackbone=Mistral-7B2024.0547.2563.94BSOBackbone=Mistral-7BBackbone=Mistral-7B2024.0543.6362.97VlguardBackbone=Mistral-7BBackbone=Mistral-7B2024.0542.8562.81Vaccine-SFTBackbone=Mistral-7BBackbone=Mistral-7B2024.0542.2546.76LisaBackbone=Mistral-7BBackbone=Mistral-7B2024.0537.5363.74