ViRA: visual agents that zoom into high-resolution photos, trained with CrossRPO on contrast sets. Models, benchmark and training data.
Huawei Lin PRO
huaweilin
AI & ML interests
None yet
Recent Activity
upvoted a collection about 19 hours ago
ViRA updated a dataset 2 days ago
huaweilin/ViRA-Train updated a dataset 2 days ago
huaweilin/ViRA-BenchOrganizations
None yet
models 9
huaweilin/ViRA-27B
Image-Text-to-Text • 27B • Updated • 69
huaweilin/ViRA-9B
Image-Text-to-Text • 9B • Updated • 71
huaweilin/VTBench_models
Updated
huaweilin/stable-diffusion-3.5-large-vae
Text-to-Image • Updated • 3
huaweilin/bsqvit_256x256
Updated • 10
huaweilin/chameleon_vqvae
Updated
huaweilin/imagenet_caption_generation
Updated
huaweilin/DMin_sd3_medium_lora_r4
Updated
huaweilin/rapidin-alpaca-llama2-7b
Updated • 6
datasets 10
huaweilin/ViRA-Train
Updated • 83
huaweilin/ViRA-Bench
Viewer • Updated • 840 • 176
huaweilin/LatComm
Viewer • Updated • 2.37M • 174 • 1
huaweilin/waste-classification
Viewer • Updated • 1.98k • 139
huaweilin/VTBench
Viewer • Updated • 50.9k • 316 • 1
huaweilin/cat_and_dog_text_to_text_image
Viewer • Updated • 30k • 14
huaweilin/image_captions_text_to_text_image
Viewer • Updated • 507k • 55
huaweilin/imagenet-1k-vl-enriched-any
Viewer • Updated • 9.31k • 5
huaweilin/DMin_mixed_datasets_8846
Viewer • Updated • 11.1k • 276
huaweilin/DMin_sd3_medium_lora_r4_caching_8846
Updated • 238