TahaKoleilat/Evi-Steer

Reproducing Results

Run the following scripts to use the checkpoints and get testing results:

(1) Few-shot evaluation setting

You can evaluate the few-shot results by downloading the models from here, placing the models in checkpoints, then running the following:

bash scripts/eval_fewshot.sh

(2) Domain generalization setting

You can evaluate the domain generalization results by downloading the models from here, placing the models in checkpoints, then running the following:

bash scripts/eval_dg.sh

Citation

If you use our work, please consider citing:

@inproceedings{koleilat2026evisteer,
  author    = {Taha Koleilat and Hassan Rivaz and Yiming Xiao},
  title     = {Evi-Steer: Learning to Steer Biomedical Vision-Language Models Through Efficient and Generalizable Evidential Tuning},
  booktitle = {Medical Image Computing and Computer Assisted Intervention -- MICCAI 2026},
  pages     = {208--218},
  year      = {2027},
  publisher = {Springer Nature Switzerland}
}