Learning Encoding-Decoding Direction Pairs to Unveil Concepts of Influence in Deep Vision Networks

ArXi:2509.23926v3 Announce Type: replace Empirical evidence shows that deep vision networks often represent concepts as directions in latent space with concept information written along directional components in the vector representation of the input. However, the mechanism to encode (write) and decode (read) concept information to and from vector representations is not directly accessible as it constitutes a latent mechanism that naturally emerges from the