List your company
ResearchFormerly WizWand

Medical Visual Question Answering on SLAKE closed-end

92.39AccuracyMEDALIGN
51.98662.475572.96583.4545Apr 19, 2024Jul 29, 2024Nov 8, 2024Feb 18, 2025May 31, 2025Sep 10, 2025Dec 21, 2025
Updated 8mo ago

Evaluation results

MethodLinks
MEDALIGNMed-LVLM=HuatuoGPT-Vision + LoRA2025.1292.39
AVISCMed-LVLM=HuatuoGPT-Vision + LoRA2025.1291.27
NucleusMed-LVLM=HuatuoGPT-Vision + LoRA2025.1290.42
GreedyMed-LVLM=HuatuoGPT-Vision + LoRA2025.1290.14
BeamMed-LVLM=HuatuoGPT-Vision + LoRA2025.1290.14
OPERAMed-LVLM=HuatuoGPT-Vision + LoRA2025.1290.14
PAIMed-LVLM=HuatuoGPT-Vision + LoRA2025.1290.14
DoLaMed-LVLM=HuatuoGPT-Vision + LoRA2025.1289.86
M3IDMed-LVLM=HuatuoGPT-Vision + LoRA2025.1289.86
VCDMed-LVLM=HuatuoGPT-Vision + LoRA2025.1289.58
DAMROMed-LVLM=HuatuoGPT-Vision + LoRA2025.1289.58
PAIMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1289.01
MEDALIGNMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1289.01
GreedyMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1288.45
BeamMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1288.45
DoLaMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1288.45
OPERAMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1288.45
Phi3.5V-MedEvaluation Protocol=Fine-Tuning, Decoding Strategy=Greedy2025.1288.19
Phi3.5V-MedEvaluation Protocol=Fine-Tuning, Decoding Strategy=VCD2025.1288.19
Phi3.5V-MedEvaluation Protocol=Fine-Tuning, Decoding Strategy=ARCD2025.1288.19
V2T-COT(Phi2)w/ MLLM=true, Act.=2.7B2025.0687.61
Phi3.5V-MedEvaluation Protocol=Fine-Tuning, Decoding Strategy=OPERA2025.1287.4
AVISCMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1287.32
M3IDMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1287.04
BiomedGPT-Mw/ MLLM=false2025.0686.8
NucleusMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1286.76
VCDMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1286.76
DAMROMed-LVLM=LLaVA-Med-1.5 + LoRA2025.1286.76
Phi3.5V-MedEvaluation Protocol=Fine-Tuning, Decoding Strategy=DoLA2025.1286.61
V2T-COT (Stablelm)w/ MLLM=true, Act.=1.6B2025.0686.48
MedThinkw/ MLLM=false, Act.=223M2025.0686.3
LLaVA-Med(LLama)w/ MLLM=true, Act.=7B2025.0685.34
LLaVA-Medtraining_setting=fully finetuned2024.0485.3
Med-MoEw/ MLLM=true, Act.=2.0B2025.0683.41
LLaVA-Med(Phi2)w/ MLLM=true, Act.=2.7B2025.0683.29
LLaVA-Med(Vicuna)w/ MLLM=true, Act.=7B2025.0683.17
PubMedCLIPw/ MLLM=false2025.0682.5
Prefix T. Medical LMw/ MLLM=true, Act.=1.5B2025.0682.01
GPT-4oEvaluation Protocol=General Domain, Decoding Strategy=Greedy2025.1274.8
HuatuoV-34BEvaluation Protocol=Zero-Shot, Decoding Strategy=Greedy2025.1274.02
Gemini Prow/ MLLM=false2025.0672.6
MOVAbackbone_parameters=7B2024.0468.8
Phi3.5V-MedEvaluation Protocol=Zero-Shot, Decoding Strategy=OPERA2025.1265.35
LLaVA-1.52024.0463.7
LLAVAw/ MLLM=true, Act.=7B2025.0663.22
Qwen-VL-7BEvaluation Protocol=General Domain, Decoding Strategy=Greedy2025.1261.42
Phi3.5V-MedEvaluation Protocol=Zero-Shot, Decoding Strategy=Greedy2025.1261.42
Phi3.5V-MedEvaluation Protocol=Zero-Shot, Decoding Strategy=VCD2025.1261.42
Phi3.5V-MedEvaluation Protocol=Zero-Shot, Decoding Strategy=ARCD2025.1261.42
Phi3.5V-MedEvaluation Protocol=Zero-Shot, Decoding Strategy=DoLA2025.1259.84
LLaVA-Med-7BEvaluation Protocol=Zero-Shot, Decoding Strategy=Greedy2025.1255.12
LLaVA-Medtraining_setting=1-epoch finetuned2024.0454.6
LLaVA-1.6-7BEvaluation Protocol=General Domain, Decoding Strategy=Greedy2025.1253.54
Phi3.5V-4.2BEvaluation Protocol=General Domain, Decoding Strategy=Greedy2025.1253.54