arxiv:2511.10615
👋 Open to Work
Anurag Pradhan
anuragpradhan
AI & ML interests
Generative models for medical imaging (MRI→CT synthesis with diffusion) · efficient VLMs/SLMs (SmolVLM2 quantization, GGUF/bnb, on-device inference) · vision transformers · AI for social good. Research intern @ IIT Mandi; final-year CS @ VIT Chennai.
Recent Activity
authored a paper 10 days ago
Towards Blind and Low-Vision Accessibility of Lightweight VLMs and Custom LLM-Evals liked a model 10 days ago
anuragpradhan/M3D-CLIP-CrossModal-MRI-CT updated a model 10 days ago
anuragpradhan/M3D-CLIP-CrossModal-MRI-CTOrganizations
None yet
models 6
anuragpradhan/M3D-CLIP-CrossModal-MRI-CT
Feature Extraction • 0.2B • Updated • 88 • 1
anuragpradhan/SmolVLM2-500M-DepthAwareVLM
Image-Text-to-Text • 0.5B • Updated • 25
anuragpradhan/SmolLM2-135M-Instruct-mlx-fp16
Text Generation • 0.1B • Updated • 10
anuragpradhan/SmolVLM2-500M-Video-Instruct-Q4_K_M-GGUF
Image-Text-to-Text • 0.4B • Updated • 108
anuragpradhan/SmolVLM2-2.2B-Instruct-bnb-4bit
Image-Text-to-Text • 0.4B • Updated • 26
anuragpradhan/SmolVLM2-500M-Video-Instruct-bnb-4bit
Image-Text-to-Text • 0.1B • Updated • 11
datasets 0
None public yet