Author

S. Mahajan

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.

#artificial intelligence Jun 2026

TEVI: Text-Conditioned Editing of Visual Representations via Sparse Autoencoders for Improved Vision-Language Alignment

This work proposes TEVI, a framework that uses captions as a signal for what to retain from image embeddings and uses sparse autoencoders to disentangle image embeddings and train a masking module to selectively reconstruct the embedding based on a given caption.

S. Mahajan, Sukrut Rao, Jiahao Xie et al. · arXiv.org · 1 citation