LINE: LLM-based Iterative Neuron Explanations for Vision Models

• URL: http://arxiv.org/abs/2604.08039v1
• Date: Thu, 09 Apr 2026 09:43:26 GMT
• Title: LINE: LLM-based Iterative Neuron Explanations for Vision Models
• Abstract summary: We introduce LINE, a training-free iterative approach tailored for open-vocabulary concept labeling in vision models.<n> LINE achieves state-of-the-art performance across multiple model architectures, yielding AUC improvements of up to 0.18 on ImageNet and 0.05 on Places365.<n> LINE provides a complete generation history, which enables polysemanticity evaluation and produces visual explanations that rival gradient-dependent activation methods.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Interpreting the concepts encoded by individual neurons in deep neural networks is a crucial step towards understanding their complex decision-making processes and ensuring AI safety. Despite recent progress in neuron labeling, existing methods often limit the search space to predefined concept vocabularies or produce overly specific descriptions that fail to capture higher-order, global concepts. We introduce LINE, a novel, training-free iterative approach tailored for open-vocabulary concept labeling in vision models. Operating in a strictly black-box setting, LINE leverages a large language model and a text-to-image generator to iteratively propose and refine concepts in a closed loop, guided by activation history. We demonstrate that LINE achieves state-of-the-art performance across multiple model architectures, yielding AUC improvements of up to 0.18 on ImageNet and 0.05 on Places365, while discovering, on average, 29% of new concepts missed by massive predefined vocabularies. Beyond identifying the top concept, LINE provides a complete generation history, which enables polysemanticity evaluation and produces supporting visual explanations that rival gradient-dependent activation maximization methods.

Related papers

• COVD: Continual Open-Vocabulary Object Detection with Novel Concept Injection [24.164864713386446]
  We introduce a new task setting, termed Continual OVD with Novel Concept Injection (COVD), where models sequentially learn incoming novel concept groups.<n>Our key observation is that pretrained visual encoders often already perceive and represent many novel concepts.<n>We propose NoIn-Det, an efficient continual injection framework without additional parameters.
  arXiv  Detail & Related papers  (2026-05-26T14:53:14Z)
• CPC-VAR:Continual Personalized and Compositional Generation in Visual Autoregressive Models [69.9266117908314]
  Visual autoregressive ( VAR) models have emerged as efficient paradigm for text-to-image generation.<n>Existing VAR-based personalization methods fail to accommodate evolving user demands.<n>We present the first systematic study of continual personalized generation in VAR models.
  arXiv  Detail & Related papers  (2026-05-19T12:18:15Z)
• Beyond Heuristic Prompting: A Concept-Guided Bayesian Framework for Zero-Shot Image Recognition [81.2779530670268]
  Vision-Language Models (VLMs) have significantly advanced zero-shot image recognition.<n>This paper enhances prompt by incorporating class-specific concepts.<n>Our method consistently outperforms state-of-the-art approaches.
  arXiv  Detail & Related papers  (2026-03-09T03:11:11Z)
• LLEXICORP: End-user Explainability of Convolutional Neural Networks [4.417922173735815]
  Concept relevance propagation (CRP) methods can backtrack predictions to these channels and find images that most activate these channels.<n>CRP are largely manual: experts must inspect activation images to name the discovered concepts and must synthesize verbose explanations from relevance maps.<n>Our approach automatically assigns descriptive names to concept prototypes and generates natural-language explanations.<n>Our findings suggest that integrating concept-based attribution methods with large language models can significantly lower the barrier to interpreting deep neural networks.
  arXiv  Detail & Related papers  (2025-11-04T16:44:45Z)
• Concept-Guided Interpretability via Neural Chunking [64.6429903327095]
  We show that neural networks exhibit patterns in their raw population activity that mirror regularities in the training data.<n>We propose three methods to extract recurring chunks on a neural population level.<n>Our work points to a new direction for interpretability, one that harnesses both cognitive principles and the structure of naturalistic data.
  arXiv  Detail & Related papers  (2025-05-16T13:49:43Z)
• OmniPrism: Learning Disentangled Visual Concept for Image Generation [57.21097864811521]
  Creative visual concept generation often draws inspiration from specific concepts in a reference image to produce relevant outcomes.<n>We propose OmniPrism, a visual concept disentangling approach for creative image generation.<n>Our method learns disentangled concept representations guided by natural language and trains a diffusion model to incorporate these concepts.
  arXiv  Detail & Related papers  (2024-12-16T18:59:52Z)
• Scaling Concept With Text-Guided Diffusion Models [53.80799139331966]
  Instead of replacing a concept, can we enhance or suppress the concept itself? We introduce ScalingConcept, a simple yet effective method to scale decomposed concepts up or down in real input without introducing new elements. More importantly, ScalingConcept enables a variety of novel zero-shot applications across image and audio domains.
  arXiv  Detail & Related papers  (2024-10-31T17:09:55Z)
• Explainable Concept Generation through Vision-Language Preference Learning for Understanding Neural Networks' Internal Representations [7.736445799116692]
  Concept-based methods have become a popular choice for explaining deep neural networks post-hoc.<n>We devise a reinforcement learning-based preference optimization algorithm that fine-tunes a vision-language generative model.<n>We demonstrate our method's ability to efficiently and reliably articulate diverse concepts.
  arXiv  Detail & Related papers  (2024-08-24T02:26:42Z)
• Discover-then-Name: Task-Agnostic Concept Bottlenecks via Automated Concept Discovery [52.498055901649025]
  Concept Bottleneck Models (CBMs) have been proposed to address the 'black-box' problem of deep neural networks. We propose a novel CBM approach -- called Discover-then-Name-CBM (DN-CBM) -- that inverts the typical paradigm. Our concept extraction strategy is efficient, since it is agnostic to the downstream task, and uses concepts already known to the model.
  arXiv  Detail & Related papers  (2024-07-19T17:50:11Z)
• Restyling Unsupervised Concept Based Interpretable Networks with Generative Models [14.604305230535026]
  We propose a novel method that relies on mapping the concept features to the latent space of a pretrained generative model.<n>We quantitatively ascertain the efficacy of our method in terms of accuracy of the interpretable prediction network, fidelity of reconstruction, as well as faithfulness and consistency of learnt concepts.
  arXiv  Detail & Related papers  (2024-07-01T14:39:41Z)
• LLM-assisted Concept Discovery: Automatically Identifying and Explaining Neuron Functions [15.381209058506078]
  Prior works have associated concepts with neurons based on examples of concepts or a pre-defined set of concepts. We propose to leverage multimodal large language models for automatic and open-ended concept discovery. We validate each concept by generating examples and counterexamples and evaluating the neuron's response on this new set of images.
  arXiv  Detail & Related papers  (2024-06-12T18:19:37Z)
• A Simple Long-Tailed Recognition Baseline via Vision-Language Model [92.2866546058082]
  The visual world naturally exhibits a long-tailed distribution of open classes, which poses great challenges to modern visual systems. Recent advances in contrastive visual-language pretraining shed light on a new pathway for visual recognition. We propose BALLAD to leverage contrastive vision-language models for long-tailed recognition.
  arXiv  Detail & Related papers  (2021-11-29T17:49:24Z)

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.