VL-SAE: Interpreting and Enhancing Vision-Language Alignment with a Unified Concept Set

• URL: http://arxiv.org/abs/2510.21323v1
• Date: Fri, 24 Oct 2025 10:29:31 GMT
• Title: VL-SAE: Interpreting and Enhancing Vision-Language Alignment with a Unified Concept Set
• Authors: Shufan Shen, Junshu Sun, Qingming Huang, Shuhui Wang,
• Abstract summary: The alignment of vision-language representations endows current Vision-Language Models with strong multi-modal reasoning capabilities.<n>We propose VL-SAE, a sparse autoencoder that encodes vision-language representations into its hidden activations.<n>For interpretation, the alignment between vision and language representations can be understood by comparing their semantics with concepts.
• Score: 80.50996301430108
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• Abstract: The alignment of vision-language representations endows current Vision-Language Models (VLMs) with strong multi-modal reasoning capabilities. However, the interpretability of the alignment component remains uninvestigated due to the difficulty in mapping the semantics of multi-modal representations into a unified concept set. To address this problem, we propose VL-SAE, a sparse autoencoder that encodes vision-language representations into its hidden activations. Each neuron in its hidden layer correlates to a concept represented by semantically similar images and texts, thereby interpreting these representations with a unified concept set. To establish the neuron-concept correlation, we encourage semantically similar representations to exhibit consistent neuron activations during self-supervised training. First, to measure the semantic similarity of multi-modal representations, we perform their alignment in an explicit form based on cosine similarity. Second, we construct the VL-SAE with a distance-based encoder and two modality-specific decoders to ensure the activation consistency of semantically similar representations. Experiments across multiple VLMs (e.g., CLIP, LLaVA) demonstrate the superior capability of VL-SAE in interpreting and enhancing the vision-language alignment. For interpretation, the alignment between vision and language representations can be understood by comparing their semantics with concepts. For enhancement, the alignment can be strengthened by aligning vision-language representations at the concept level, contributing to performance improvements in downstream tasks, including zero-shot image classification and hallucination elimination. Codes are available at https://github.com/ssfgunner/VL-SAE.

Related papers

• LUCID-SAE: Learning Unified Vision-Language Sparse Codes for Interpretable Concept Discovery [14.222802170483739]
  LUCID is a vision-language sparse autoencoder that learns a shared latent dictionary for image patch and text token representations.<n> LUCID yields interpretable shared features that support patch-level grounding, establish cross-modal neuron correspondence, and enhance robustness against the concept clustering problem.<n>Our analysis reveals that LUCID's shared features capture diverse semantic categories beyond objects, including actions, attributes, and abstract concepts.
  arXiv  Detail & Related papers  (2026-02-07T02:01:25Z)
• Youtu-VL: Unleashing Visual Potential via Unified Vision-Language Supervision [79.06371915084833]
  We introduce Youtu-VL, a framework leveraging the Vision-Language Unified Autoregressive Supervision (VLUAS) paradigm.<n>Youtu-VL applies unified autoregressive supervision to both visual details and linguistic content.<n>We extend this paradigm to encompass vision-centric tasks, enabling a standard VLM to perform vision-centric tasks without task-specific additions.
  arXiv  Detail & Related papers  (2026-01-27T17:01:16Z)
• CLMN: Concept based Language Models via Neural Symbolic Reasoning [27.255064617527328]
  Concept Language Model Network (CLMN) is a neural-symbolic framework that keeps both performance and interpretability.<n> CLMN represents concepts as continuous, human-readable embeddings.<n>Model augments original text features with concept-aware representations and automatically induces interpretable logic rules.
  arXiv  Detail & Related papers  (2025-10-11T06:58:44Z)
• Improving vision-language alignment with graph spiking hybrid Networks [10.88584928028832]
  This paper proposes a comprehensive visual semantic representation module, necessitating the utilization of panoptic segmentation to generate fine-grained semantic features.<n>We propose a novel Graph Spiking Hybrid Network (GSHN) that integrates the complementary advantages of Spiking Neural Networks (SNNs) and Graph Attention Networks (GATs) to encode visual semantic information.
  arXiv  Detail & Related papers  (2025-01-31T11:55:17Z)
• VladVA: Discriminative Fine-tuning of LVLMs [67.14293827774827]
  Contrastively-trained Vision-Language Models (VLMs) like CLIP have become the de facto approach for discriminative vision-language representation learning.<n>We propose to combine "the best of both worlds": a new training approach for discriminative fine-tuning of LVLMs.
  arXiv  Detail & Related papers  (2024-12-05T17:54:27Z)
• 3VL: Using Trees to Improve Vision-Language Models' Interpretability [40.678288227161936]
  Vision-Language models (VLMs) have proven to be effective at aligning image and text representations, producing superior zero-shot results when transferred to many downstream tasks.<n>These representations suffer from some key shortcomings in understanding Compositional Language Concepts (CLC), such as recognizing objects' attributes, states, and relations between different objects.<n>In this work, we introduce the architecture and training technique of Tree-augmented Vision-Language (3VL) model accompanied by our proposed Anchor inference method and Differential Relevance (DiRe) interpretability tool.
  arXiv  Detail & Related papers  (2023-12-28T20:26:03Z)
• Identifying Interpretable Subspaces in Image Representations [54.821222487956355]
  We propose a framework to explain features of image representations using Contrasting Concepts (FALCON) For a target feature, FALCON captions its highly activating cropped images using a large captioning dataset and a pre-trained vision-language model like CLIP. Each word among the captions is scored and ranked leading to a small number of shared, human-understandable concepts.
  arXiv  Detail & Related papers  (2023-07-20T00:02:24Z)
• Progressive Semantic-Visual Mutual Adaption for Generalized Zero-Shot Learning [74.48337375174297]
  Generalized Zero-Shot Learning (GZSL) identifies unseen categories by knowledge transferred from the seen domain. We deploy the dual semantic-visual transformer module (DSVTM) to progressively model the correspondences between prototypes and visual features. DSVTM devises an instance-motivated semantic encoder that learns instance-centric prototypes to adapt to different images, enabling the recast of the unmatched semantic-visual pair into the matched one.
  arXiv  Detail & Related papers  (2023-03-27T15:21:43Z)
• DiMBERT: Learning Vision-Language Grounded Representations with Disentangled Multimodal-Attention [101.99313208598569]
  Vision-and-language (V-L) tasks require the system to understand both vision content and natural language. We propose DiMBERT (short for Disentangled Multimodal-Attention BERT), which applies separated attention spaces for vision and language. We show that DiMBERT sets new state-of-the-art performance on three tasks.
  arXiv  Detail & Related papers  (2022-10-28T23:00:40Z)
• Single-Stream Multi-Level Alignment for Vision-Language Pretraining [103.09776737512078]
  We propose a single stream model that aligns the modalities at multiple levels. We achieve this using two novel tasks: symmetric cross-modality reconstruction and a pseudo-labeled key word prediction. We demonstrate top performance on a set of Vision-Language downstream tasks such as zero-shot/fine-tuned image/text retrieval, referring expression, and VQA.
  arXiv  Detail & Related papers  (2022-03-27T21:16:10Z)

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.