mlpc-ucsd / BLIVA

(AAAI 2024) BLIVA: A Simple Multimodal LLM for Better Handling of Text-rich Visual Questions

https://arxiv.org/abs/2308.09936

mlpc-ucsd/BLIVA Issues

Error "input lengths of input ids is 0"
Updated a month ago2
Question about zero-shot results on general VQA benchmarks
Closed 10 months ago5
performance on the VizWiz dataset
Updated 6 months ago6
Couldn't run `evaluate.py`
Updated 6 months ago3
Can't download the model weights
Closed 7 months ago3
could you provide the dataset code for visdial and the test code?
Closed 7 months ago12
Custom Data and Internet Connection for LLM
Updated 7 months ago
Inquire if image captioning tasks are supported
Updated 8 months ago4
Some question about train dataset？
Closed 9 months ago1
Cuda Out of Memory!
Updated 9 months ago2
how to split triain, val and test set of ocr-vqa dataset?
Updated 9 months ago2
final loss of pretrain stage
Updated 9 months ago
Training datasets
Closed 10 months ago2
"ValueError: Attempting to unscale FP16 gradients" in vicuna_v1.1
Updated 10 months ago2
reduction="none" problems
Closed 10 months ago1
Is Video Supported?
Updated 10 months ago1
Traing code
Updated 10 months ago1
Vram?
Closed a year ago6