arXiv AI June 24, 2026 By Jonas Klotz, Cassio F. Dantas, Pallavi Jain, Diego Marcos, Beg\"um Demir

Evaluating the Interpretability of Sparse Autoencoders with Concept Annotations

arXiv:2606. 24716v1 Announce Type: cross Abstract: Sparse autoencoders (SAEs) are increasingly used to extract interpretable concepts from vision and vision language models, yet existing evaluation methods largely rely on proxy metrics or qualitative inspection rather than measuring semantic correspondence.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv AI.

arXiv AI

Jul 10

When Structured Sparse Autoencoders Learn Consistent Concepts Across Modalities

arXiv:2607. 08605v1 Announce Type: cross Abstract: 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.

By Weiduo Liao, Yunqiao Yang, Ying Wei

llms efficiency multimodal safety

arXiv Machine Learning

Aug 12

Clarity: The Flexibility-Interpretability Trade-Off in Sparsity-aware Concept Bottleneck Models

arXiv:2601. 21944v3 Announce Type: replace Abstract: The widespread adoption of deep learning models in computer vision has intensified concerns about interpretability.

By Konstantinos P. Panousis, Diego Marcos

computer-vision efficiency multimodal safety

arXiv AI

Jul 7

TORINO: Token Reduction via Interpretable Concept Overlap in Vision-Language Models

arXiv:2607. 04593v1 Announce Type: cross Abstract: Vision-Language Models (VLMs) have demonstrated impressive capabilities across different tasks, but their computational cost is dominated by the large number of visual tokens fed to the language model.

By Riccardo Renzulli, Gabriele Spadaro, Shruthi Gowda, Alaa Eddine Mazouz, Van-Tam Nguyen

llms diffusion fine-tuning efficiency multimodal benchmarks

arXiv AI

Jun 16

Beyond Scalar Distances: Semantic Attribute Gradients from Frozen MLLMs for Visual Embeddings

arXiv:2606. 15134v1 Announce Type: cross Abstract: Vision encoders for retrieval are typically trained with class-label supervision: each training pair reduces to a scalar that uniformly pushes the embedding apart or pulls it together, as if every visual attribute either differed or matched.

By Shubhang Bhatnagar, Dheeraj Baiju, Narendra Ahuja

llms rag efficiency multimodal benchmarks

arXiv AI

Jun 17

Plug-and-Adapt: Multimodal Coreference Resolution at First Sight with a Pretrained Alignment Model

arXiv:2606. 17950v1 Announce Type: cross Abstract: Visual information helps resolve ambiguity in coreference resolution, leading to notable performance gains.

By Jinghan Wu, Jing Li, Ivor W. Tsang, Xuetao Zhang

multimodal benchmarks safety

arXiv AI

Jul 21

Measuring Monosemanticity in Sparse Autoencoders via Latent Activation Coherence

arXiv:2607. 17770v1 Announce Type: cross Abstract: Within Explainable Artificial Intelligence, mechanistic interpretability uses Sparse Autoencoders (SAEs) to extract more interpretable features from neural representations.

By Katarzyna Filus, Sebastian Pokuci\'nski

llms rag efficiency multimodal safety

Related stories

When Structured Sparse Autoencoders Learn Consistent Concepts Across Modalities

Clarity: The Flexibility-Interpretability Trade-Off in Sparsity-aware Concept Bottleneck Models

TORINO: Token Reduction via Interpretable Concept Overlap in Vision-Language Models

Beyond Scalar Distances: Semantic Attribute Gradients from Frozen MLLMs for Visual Embeddings

Plug-and-Adapt: Multimodal Coreference Resolution at First Sight with a Pretrained Alignment Model

Measuring Monosemanticity in Sparse Autoencoders via Latent Activation Coherence