CORTEXA
← Browse
arxivcs.CVcs.AIcs.LG2026-07-09

When Structured Sparse Autoencoders Learn Consistent Concepts Across Modalities

Weiduo Liao, Yunqiao Yang, Ying Wei

Sparse autoencoders (SAEs) have emerged as a promising technique for mechanistic interpretability by learning a set of sparse latent features in large models, each of which encodes a distinct concept. However, in vision-language models (VLMs), vanilla SAEs struggle to learn modality-consistent concepts, with concepts often exhibiting fragmented coverage (i.e., disjoint regions) in the visual modality. To address this challenge, we propose a Structured Sparse AutoEncoder ($S^2AE$) that enforces concept consistency from both semantic and spatial perspectives in the visual modality. Specifically, we group image patches based on Transformer attention similarity and spatial proximity, and introduce a structured sparsity regularization when training the vanilla SAE. The regularization consists of exclusive sparsity for inter-group concept disentanglement and group sparsity for intra-group concept consistency, which drives the latent neurons by SAEs to specialize in distinct, semantically grounded concepts. Evaluated on the \texttt{Qwen2.5-VL-7B-Instruct} model, the method achieves 6.06% average improvement in semantic alignment (mIoU) and 60.81 in representational efficiency (lower l0 norm) while maintaining near-perfect reconstruction fidelity with an Explained Variance above 99%. Cross-modal analysis further demonstrates that $S^2AE$ enhances neuronal monosemanticity by this visual structural prior, achieving a 3.08% average gain in semantic consistency and a 2.37% average gain in monosemanticity scores for both modalities of multimodal features, thereby fostering more coherent and disentangled representations.

View free PDFSource page

Related papers

arxivcs.CVcs.AIcs.LG2026-07-20

Measuring Monosemanticity in Sparse Autoencoders via Latent Activation Coherence

Katarzyna Filus, Sebastian Pokuciński

Within Explainable Artificial Intelligence, mechanistic interpretability uses Sparse Autoencoders (SAEs) to extract more interpretable features from neural representations. However, assessing their monosemanticity, and thus explanation quality, remains challenging. Existing metri…

View free PDFSource page
arxivcs.CVcs.AIcs.LG2026-07-07

Dynamic-in-Few-Step: Unifying Dynamic Computation and Few-Step Distillation for Efficient Video Generation

Yu Cheng, Siyue Yao, Zhongang Qi, Shanyan Guan, Wei Li, Fajie Yuan

Video Diffusion Models (VDMs) have demonstrated superior generation quality but suffer from prohibitive computational costs. While recent few-step distillation techniques significantly accelerate inference, they typically enforce a static model architecture across all denoising s…

View free PDFSource page
arxivcs.CVcs.AIcs.LG2026-06-29

Set-Inclusive Uncertainty Modeling for Robust Brain Tumor Segmentation

Seunghun Baek, Jihwan Park, Jaeyoon Sim, Hoseok Lee, Seungjoo Lee, Won Hwa Kim

Multimodal MRI is essential for accurate brain tumor segmentation. However, acquiring all modalities at inference is often challenging in practice, which causes intrinsic uncertainty due to unavoidable information loss. Without modeling this uncertainty, existing methods encode i…

View free PDFSource page
arxivcs.CVcs.AIcs.LG2026-07-06

Graph Representation Learning of Longitudinal Medical Imaging Trajectories for Treatment Response Prediction

Johannes Kiechle, Richard Osuala, Daniel M. Lang, Stefan M. Fischer, Ivana Janíčková, Karim Lekadir, et al.

In patients with breast cancer, pathological complete response (pCR) has been established as a clinically meaningful surrogate marker for long-term outcomes. While commonly treated with neoadjuvant chemotherapy (NACT), effective treatment decision-making remains challenging, as t…

View free PDFSource page
arxivcs.LGcs.AIcs.CVcs.DC2026-07-13

Continual Learning with Elastic Regularization and Synthetic Replay for Federated MLLM Fine-Tuning

Jing Liu, Chenxuanyin Zou, Jiayang Ren, Gaoyun Fang, Chengfang Li, Yan Wang, et al.

Federated fine-tuning of Multimodal Large Language Models (MLLMs) across distributed networks enables privacy-sensitive adaptation to evolving data streams, yet a fundamental obstacle prevents robust deployment in dynamic environments: catastrophic forgetting, wherein sequential…

View free PDFSource page