Same Concept, Different Directions: Cross-Modal Feature Heterogeneity in Sparse Autoencoders

Chungpa Lee, Jihoon Kwon, Kyle Min, Jy-yong Sohn

Keywords: Concept Discovery (e.g., SAEs, dictionary learning), Feature Geometry, Methods (probing, steering, causal interventions)

TL;DR: The same concept can occupy different directions across modalities, and preserving these directions enables better cross-modal control through post hoc alignment.

Abstract: Vision-language models map images and text into a joint embedding space. However, these embeddings often entangle multiple semantic features, which limits their interpretability and controllability. While sparse autoencoders have emerged as a useful tool for decomposing these embeddings into monosemantic features, their application to joint embedding spaces has largely relied on an implicit, untested assumption that semantically corresponding features share the same directions across modalities. In this paper, we challenge this assumption by identifying discrepancies in feature directions for the same concept across image and text modalities, a phenomenon we term cross-modal feature heterogeneity. We demonstrate that this heterogeneity is a key driver of the modality split, where a shared concept activates different latents depending on the modality. This finding further reveals why aligning latent activations alone is insufficient to resolve the underlying feature mismatch. To address this misalignment, we propose an approach that trains sparse autoencoders to preserve the unique feature geometry of each modality and aligns corresponding features post hoc. Our method improves reconstruction fidelity and enhances performance in cross-modal retrieval and concept steering.

Submission Number: 131