Visual Exploration of Feature Relationships in Sparse Autoencoders with Curated Concepts

• URL: http://arxiv.org/abs/2511.06048v1
• Date: Sat, 08 Nov 2025 15:36:57 GMT
• Title: Visual Exploration of Feature Relationships in Sparse Autoencoders with Curated Concepts
• Authors: Xinyuan Yan, Shusen Liu, Kowshik Thopalli, Bei Wang,
• Abstract summary: We propose a focused exploration framework that prioritizes curated concepts and their corresponding SAE features over attempts to visualize all available features simultaneously.<n>We present an interactive visualization system that combines topology-based visual encoding with dimensionality reduction to faithfully represent both local and global relationships among selected features.
• Score: 8.768503486874623
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Sparse autoencoders (SAEs) have emerged as a powerful tool for uncovering interpretable features in large language models (LLMs) through the sparse directions they learn. However, the sheer number of extracted directions makes comprehensive exploration intractable. While conventional embedding techniques such as UMAP can reveal global structure, they suffer from limitations including high-dimensional compression artifacts, overplotting, and misleading neighborhood distortions. In this work, we propose a focused exploration framework that prioritizes curated concepts and their corresponding SAE features over attempts to visualize all available features simultaneously. We present an interactive visualization system that combines topology-based visual encoding with dimensionality reduction to faithfully represent both local and global relationships among selected features. This hybrid approach enables users to investigate SAE behavior through targeted, interpretable subsets, facilitating deeper and more nuanced analysis of concept representation in latent space.

Related papers

• Revisiting Multi-Task Visual Representation Learning [52.93947931352643]
  We introduce MTV, a principled multi-task visual pretraining framework.<n>We leverage high-capacity "expert" models to synthesize dense, structured pseudo-labels at scale.<n>Our results demonstrate that MTV achieves "best-of-both-worlds" performance.
  arXiv  Detail & Related papers  (2026-01-20T11:59:19Z)
• Multi-label Classification with Panoptic Context Aggregation Networks [61.82285737410154]
  This paper introduces the Deep Panoptic Context Aggregation Network (PanCAN), a novel approach that hierarchically integrates multi-order geometric contexts.<n>PanCAN learns multi-order neighborhood relationships at each scale by combining random walks with an attention mechanism.<n>Experiments on NUS-WIDE, PASCAL VOC,2007, and MS-COCO benchmarks demonstrate that PanCAN consistently achieves competitive results.
  arXiv  Detail & Related papers  (2025-12-29T14:16:21Z)
• Contextualized Representation Learning for Effective Human-Object Interaction Detection [17.242400169885453]
  Human-Object Interaction (HOI) detection aims to simultaneously localize human-object pairs and recognize their interactions.<n>We introduce a Contextualized Representation Learning that integrates both affordance-guided reasoning and contextual prompts.<n>Our proposed method demonstrates superior performance on both the HICO-Det and V-COCO datasets in most scenarios.
  arXiv  Detail & Related papers  (2025-09-16T08:03:16Z)
• Generalized Decoupled Learning for Enhancing Open-Vocabulary Dense Perception [71.26728044621458]
  DeCLIP is a novel framework that enhances CLIP by decoupling the self-attention module to obtain content'' and context'' features respectively.<n>It consistently achieves state-of-the-art performance across a broad spectrum of tasks, including 2D detection and segmentation, 3D instance segmentation, video instance segmentation, and 6D object pose estimation.
  arXiv  Detail & Related papers  (2025-08-15T06:43:51Z)
• Collaborative Perceiver: Elevating Vision-based 3D Object Detection via Local Density-Aware Spatial Occupancy [7.570294108494611]
  Vision-based bird's-eye-view (BEV) 3D object detection has advanced significantly in autonomous driving.<n>Existing methods often construct 3D BEV representations by collapsing extracted object features.<n>We introduce a multi-task learning framework, Collaborative Perceiver, to bridge gaps in spatial representations and feature refinement.
  arXiv  Detail & Related papers  (2025-07-28T21:56:43Z)
• MapExplorer: New Content Generation from Low-Dimensional Visualizations [60.02149343347818]
  Low-dimensional visualizations, or "projection maps," are widely used to interpret large-scale and complex datasets.<n>These visualizations not only aid in understanding existing knowledge spaces but also implicitly guide exploration into unknown areas.<n>We introduce MapExplorer, a novel knowledge discovery task that translates coordinates within any projection map into coherent, contextually aligned textual content.
  arXiv  Detail & Related papers  (2024-12-24T20:16:13Z)
• Localization, balance and affinity: a stronger multifaceted collaborative salient object detector in remote sensing images [24.06927394483275]
  We propose a stronger multifaceted collaborative salient object detector in ORSIs, termed LBA-MCNet. The network focuses on accurately locating targets, balancing detailed features, and modeling image-level global context information.
  arXiv  Detail & Related papers  (2024-10-31T14:50:48Z)
• Disentangling Dense Embeddings with Sparse Autoencoders [0.0]
  Sparse autoencoders (SAEs) have shown promise in extracting interpretable features from complex neural networks. We present one of the first applications of SAEs to dense text embeddings from large language models. We show that the resulting sparse representations maintain semantic fidelity while offering interpretability.
  arXiv  Detail & Related papers  (2024-08-01T15:46:22Z)
• Embedding Generalized Semantic Knowledge into Few-Shot Remote Sensing Segmentation [26.542268630980814]
  Few-shot segmentation (FSS) for remote sensing (RS) imagery leverages supporting information from limited annotated samples to achieve query segmentation of novel classes. Previous efforts are dedicated to mining segmentation-guiding visual cues from a constrained set of support samples. We propose a holistic semantic embedding (HSE) approach that effectively harnesses general semantic knowledge.
  arXiv  Detail & Related papers  (2024-05-22T14:26:04Z)
• ZoomNeXt: A Unified Collaborative Pyramid Network for Camouflaged Object Detection [70.11264880907652]
  Recent object (COD) attempts to segment objects visually blended into their surroundings, which is extremely complex and difficult in real-world scenarios. We propose an effective unified collaborative pyramid network that mimics human behavior when observing vague images and camouflaged zooming in and out. Our framework consistently outperforms existing state-of-the-art methods in image and video COD benchmarks.
  arXiv  Detail & Related papers  (2023-10-31T06:11:23Z)
• Weakly-supervised Contrastive Learning for Unsupervised Object Discovery [52.696041556640516]
  Unsupervised object discovery is promising due to its ability to discover objects in a generic manner. We design a semantic-guided self-supervised learning model to extract high-level semantic features from images. We introduce Principal Component Analysis (PCA) to localize object regions.
  arXiv  Detail & Related papers  (2023-07-07T04:03:48Z)
• Robust Saliency-Aware Distillation for Few-shot Fine-grained Visual Recognition [57.08108545219043]
  Recognizing novel sub-categories with scarce samples is an essential and challenging research topic in computer vision. Existing literature addresses this challenge by employing local-based representation approaches. This article proposes a novel model, Robust Saliency-aware Distillation (RSaD), for few-shot fine-grained visual recognition.
  arXiv  Detail & Related papers  (2023-05-12T00:13:17Z)
• Closed-Form Factorization of Latent Semantics in GANs [65.42778970898534]
  A rich set of interpretable dimensions has been shown to emerge in the latent space of the Generative Adversarial Networks (GANs) trained for synthesizing images. In this work, we examine the internal representation learned by GANs to reveal the underlying variation factors in an unsupervised manner. We propose a closed-form factorization algorithm for latent semantic discovery by directly decomposing the pre-trained weights.
  arXiv  Detail & Related papers  (2020-07-13T18:05:36Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.