List your company
ResearchFormerly WizWand

Open-ended Visual Question Answering on PMC-VQA (test-initial)

35.4BLEU-1MedVInT-TE
21.98425.46728.9532.433May 17, 2023
Updated 8mo ago

Evaluation results

MethodLinks
MedVInT-TELanguage Backbone=LLAMA-ENC [54], Vision Backbone=PMC-CLIP [32], Evaluation Protocol=Trained on PMC-VQA2023.0535.418.2
MedVInT-TELanguage Backbone=PubMedBERT [20], Vision Backbone=PMC-CLIP [32], Evaluation Protocol=Trained on PMC-VQA2023.0535.222
MedVInT-TELanguage Backbone=PMC-LLAMA-ENC [57], Vision Backbone=PMC-CLIP [32], Evaluation Protocol=Trained on PMC-VQA2023.0534.818.1
MedVInT-TELanguage Backbone=PubMedBERT [20], Vision Backbone=Scratch, Evaluation Protocol=Trained on PMC-VQA2023.0533.720.4
MedVInT-TELanguage Backbone=PubMedBERT [20], Vision Backbone=CLIP [47], Evaluation Protocol=Trained on PMC-VQA2023.0533.720.4
MedVInT-TELanguage Backbone=PMC-LLAMA-ENC [57], Vision Backbone=CLIP [47], Evaluation Protocol=Trained on PMC-VQA2023.053316.6
MedVInT-TDLanguage Backbone=PMC-LLAMA [57], Vision Backbone=PMC-CLIP [32], Evaluation Protocol=Trained on PMC-VQA2023.0532.720.3
MedVInT-TELanguage Backbone=PMC-LLAMA-ENC [57], Vision Backbone=Scratch, Evaluation Protocol=Trained on PMC-VQA2023.0532.616.2
MedVInT-TELanguage Backbone=LLAMA-ENC [54], Vision Backbone=Scratch, Evaluation Protocol=Trained on PMC-VQA2023.0532.515.3
MedVInT-TELanguage Backbone=LLAMA-ENC [54], Vision Backbone=CLIP [47], Evaluation Protocol=Trained on PMC-VQA2023.0532.315.6
MedVInT-TDLanguage Backbone=LLAMA [54], Vision Backbone=PMC-CLIP [32], Evaluation Protocol=Trained on PMC-VQA2023.0531.920
MedVInT-TDLanguage Backbone=PMC-LLAMA [57], Vision Backbone=CLIP [47], Evaluation Protocol=Trained on PMC-VQA2023.0531.419.5
MedVInT-TDLanguage Backbone=LLAMA [54], Vision Backbone=CLIP [47], Evaluation Protocol=Trained on PMC-VQA2023.0531.319.5
MedVInT-TDLanguage Backbone=LLAMA [54], Vision Backbone=Scratch, Evaluation Protocol=Trained on PMC-VQA2023.0529.117.4
MedVInT-TDLanguage Backbone=PMC-LLAMA [57], Vision Backbone=Scratch, Evaluation Protocol=Trained on PMC-VQA2023.0528.616.8
LLaVA-MedLanguage Backbone=Vicuna [14], Vision Backbone=BioMedCLIP [64], Evaluation Protocol=Zero-shot2023.0528.33.7
Open-FlamingoLanguage Backbone=LLAMA [54], Vision Backbone=CLIP [47], Evaluation Protocol=Zero-shot2023.0526.14.1
LLAMALanguage Backbone=LLAMA [54], Vision Backbone=None, Evaluation Protocol=Trained on PMC-VQA2023.0526.114.2
BLIP-2Language Backbone=OPT-2.7B [63], Vision Backbone=CLIP [47], Evaluation Protocol=Zero-shot2023.0522.55.2