MS student, EPFL - EPF Lausanne
2 papers at NeurIPS 2025
MEIcoder uses most exciting inputs, SSIM loss, and adversarial training to reconstruct high-fidelity visual stimuli from neural population activity, excelling in data- and neuron-scarce settings.
This paper introduces a novel scenario, weak-to-strong generalization under distribution shifts, and proposes Robust AdaptiVe wEightiNg (RAVEN) to tackle this challenge.
