SCAM: A Real-World Typographic Robustness Evaluation for Multimodal Foundation Models
Paper • 2504.04893 • Published
Ju
Juicstus
AI & ML interests
bio-inspired (NeuroAI), applied foundation models (intersection of digital and non-digital world), continual learning
Recent Activity
authored a paper 1 day ago
SCAM: A Real-World Typographic Robustness Evaluation for Multimodal
Foundation Models authored a paper 1 day ago
Programs-of-Layers in LLMs through the Lens of Cortical Areas commentedon a paper 6 days ago
MemT-1.0: Towards Memory-Native EvolutionOrganizations
authored 2
papers 1 day ago
commented a
paper 6 days ago
liked a
dataset 6 days ago
published a
dataset 6 days ago
updated a
dataset 6 days ago
Images cropped?
1
#2 opened 5 months ago
by Juicstus
Add SWE-Bench Pro evaluation results
1
#46 opened 9 months ago
by 
nielsr
Regarding the calculation of VRAM requirements for deploying meta-llama/Llama-4-Maverick-17B-128E-Instruct
1
#45 opened about 1 year ago
by 
a58982284
Ahead of its time
#1 opened 6 months ago
by Juicstus
updated 2
datasets 6 months ago
updated a
dataset 7 months ago
Importants of this model
1
#2 opened 9 months ago
by 
0xwumade
liked a
dataset 10 months ago
minor adjustments
#1 opened 10 months ago
by 
Juicstus
minor changes
#3 opened 10 months ago
by Juicstus
Update README.md
1
#1 opened 11 months ago
by Juicstus
liked a
dataset 11 months ago
liked a
Space 12 months ago