[EMNLP'25] A Benchmark for Assessing VLM Safety with Real-World Memes
DongGeon Lee
oneonlee
AI & ML interests
Data-centric natural language processing, AI Safety
Recent Activity
upvoted a paper 1 day ago
Scaling Synthetic Data Creation with 1,000,000,000 Personas liked a dataset 1 day ago
proj-persona/PersonaHub upvoted a paper 3 months ago
What Users Leave Unsaid: Under-Specified Queries Limit Vision-Language ModelsOrganizations
None yet
models 14
oneonlee/llama-3.1-nemoguard-8b-content-safety-merged
8B • Updated • 290
oneonlee/KorMath-Llama-3.1-8B-SFT
Text Generation • Updated
oneonlee/Llama-3.1-8B-SFT
Text Generation • 8B • Updated
oneonlee/KorMath-Llama-3.1-8B
Text Generation • 8B • Updated
oneonlee/Unverified-KorMath-Llama-3.2-1B-SFT
1B • Updated
oneonlee/Unverified-KorMath-Llama-3.2-1B
Text Generation • 1B • Updated
oneonlee/KorMath-Llama-3.2-3B-SFT
Text Generation • 3B • Updated
oneonlee/KorMath-Llama-3.2-1B-SFT
Text Generation • 1B • Updated
oneonlee/Llama-3.2-1B-SFT
Text Generation • 1B • Updated
oneonlee/KorMath-Llama-3.2-3B
Text Generation • 3B • Updated
datasets 7
oneonlee/SIUO
Viewer • Updated • 334 • 10
oneonlee/Meme-Safety-Bench
Viewer • Updated • 50.8k • 98 • 5
oneonlee/aegis2-train
Viewer • Updated • 30k • 5
oneonlee/wildguardmix-train
Viewer • Updated • 86.8k • 9
oneonlee/Wiki-NFQA
Viewer • Updated • 1.89k • 4 • 1
oneonlee/wiki-kg-prob
Viewer • Updated • 230 • 2
oneonlee/cleansed_emocontext
Viewer • Updated • 38.4k • 51 • 2