arXiv Machine Learning

Do Sparse Autoencoders Learn Meaningful Concept Hierarchies?

arXiv AI

Jun 16

Cascaded Sparse Autoencoders Learn Multi-Level Visual Concepts in Multimodal LLMs

arXiv:2606. 16193v1 Announce Type: cross Abstract: Multimodal Large Language Models (MLLMs) have demonstrated strong performance on vision-language tasks, yet their internal visual representations remain difficult to interpret.

By Yusong Zhao, Hengyi Wang, Tanuja Ganu, Akshay Nambi, Hao Wang

llms multimodal benchmarks safety

arXiv AI

1d ago

SAE++: Cascaded Sparse Autoencoders Learn Multi-Level Visual Concepts in Multimodal LLMs

arXiv:2606.16193v2 Announce Type: replace-cross Abstract: Multimodal Large Language Models (MLLMs) have demonstrated strong performance on vision-language tasks, yet their internal visual representat...

By Yusong Zhao, Hengyi Wang, Tanuja Ganu, Akshay Nambi, Hao Wang

llms multimodal benchmarks safety

arXiv AI

Jun 9

A Geometric Unification of Concept Learning with Concept Cones

arXiv:2512. 07355v2 Announce Type: replace Abstract: Two traditions of interpretability have evolved side by side but seldom spoken to each other: Concept Bottleneck Models (CBMs), which prescribe what a concept should be, and Sparse Autoencoders (SAEs), which discover what concepts emerge.

By Alexandre Rocchi, Thomas Fel, Gianni Franchi

efficiency safety

arXiv Computer Vision

Oct 1

Towards Open-Ended Visual Scientific Discovery with Sparse Autoencoders

arXiv:2511.17735v2 Announce Type: replace Abstract: Foundation models in several scientific domains, including visual domains, learn representations that capture complex semantics from their respecti...

By Jacob Beattie, Samuel Stevens, Neil Rosser, Yu Su, Tanya Berger-Wolf

arXiv Machine Learning

Jun 5

Concept-SAE: A Controllable and Invertible Concept Interface for Sparse Autoencoders

arXiv:2509. 22015v2 Announce Type: replace Abstract: Standard Sparse Autoencoders (SAEs) excel at discovering a dictionary of a model's learned features, providing a powerful lens for passive feature discovery.

By Jianrong Ding, Muxi Chen, Chenchen Zhao, Qiang Xu

safety

arXiv AI

Jul 2

Identifying Latent Concepts and Structures for Generalized Category Discovery

arXiv:2607. 00620v1 Announce Type: cross Abstract: Generalized Category Discovery (GCD) aims to recognize known classes while autonomously discovering novel ones in open-world settings.

By Boyang Dai, Chaoqi Chen, Yizhou Yu

rag safety

arXiv AI

Jul 7

Explainable Novel Category Discovery in Semantic Concept Space

arXiv:2607. 04548v1 Announce Type: cross Abstract: Novel category discovery aims to identify unseen classes from unlabeled data by transferring knowledge from labeled categories, but most existing methods perform discovery in opaque latent feature spaces.

By Ifrat Ikhtear Uddin, Yang Zhou, KC Santosh, Longwei Wang

multimodal

arXiv Machine Learning

Aug 12

Beyond a Bag of Features: Set-Level Instability in Sparse Autoencoders

arXiv:2608. 11197v1 Announce Type: new Abstract: Shani et al.

By Nikolai Bolik, Lennart St\"opler, Artur Andrzejak

llms rag

arXiv Machine Learning

Sep 30

When Trees Are Not Enough: Learning Mixed-Topology Feature Graphs with Adaptive Graph Sparse Autoencoders

arXiv:2609.36294v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) expose interpretable features in large language model activations, yet existing structured SAEs impose single-parent trees o...

By Xiaozuo Shen, Yifei Cai, Tian Tan, Rui Ning, Chunsheng Xin, Hongyi Wu

llms

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

Jul 24

Interpretable Embeddings with Sparse Autoencoders: A Data Analysis Toolkit

arXiv:2512. 10092v2 Announce Type: replace Abstract: Analyzing large-scale text corpora is a core challenge in machine learning, crucial for tasks like identifying undesirable model behaviors or biases in training data.

By Nick Jiang, Xiaoqing Sun, Lisa Dunlap, Lewis Smith, Neel Nanda

llms rag

arXiv AI

Sep 1

Enhancing SAE-based Steering via Neighbor Integrated Feature Selection

arXiv:2608.28806v1 Announce Type: new Abstract: Sparse autoencoders (SAEs) disentangle model activations into interpretable features and are widely used for steering large language models. Most exist...

By Yutian Liu, Xu Wang, Difan Zou

llms

Related stories