Author

Mohamed Ansaff

1 paper indexed here

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Preprint Oct 2026

M-PFT: Angular Margin for Calibrated Prompt-Tuning of Fine-Tuned Vision-Language Models

Prompt-tuning has emerged as a parameter-efficient technique for adapting large-scale vision-language models (VLMs) to downstream tasks while keeping the pre-trained weights frozen. However, prompt-tuned VLMs often suffer from a base-to-novel dual miscalibration problem, where models typically show underconfidence on b...

Shihab Aaqil Ahamed, Mohamed Ansaff, Nang Khin Myat Lay Phyu et al. · 0 citations