CAMAL: Improving Attention Alignment and Faithfulness with Segmentation Masks

• URL: http://arxiv.org/abs/2605.08325v1
• Date: Fri, 08 May 2026 16:51:52 GMT
• Title: CAMAL: Improving Attention Alignment and Faithfulness with Segmentation Masks
• Abstract summary: We propose Class Activation Map Attention Learning (CAMAL) to improve attention alignment and faithfulness in vision models.<n>CAMAL extracts the model's attention for each image during training and compares attention to ground-truth discriminative regions obtained from the corresponding segmentation masks.<n>CAMAL acts as an auxiliary regularizer, encouraging attention that aligns with ground-truth discriminative regions, while suppressing attention elsewhere.
• Score: 20.876949110882325
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Many vision datasets now provide segmentation masks in addition to annotated images to support a wide range of tasks. In this work, we propose Class Activation Map Attention Learning (CAMAL), an efficient and scalable method that utilizes segmentation masks to improve attention alignment and faithfulness in vision models. Specifically, attention alignment refers to the degree to which a model's attention aligns with ground-truth discriminative regions, while attention faithfulness refers to the degree to which a model's attention influences its decision. Improving both attention alignment and faithfulness is essential for ensuring that model attention is both spatially accurate and causally meaningful. To improve attention alignment and faithfulness in vision models, CAMAL first extracts the model's attention for each image during training and then compares the attention to ground-truth discriminative regions obtained from the corresponding segmentation masks. CAMAL then acts as an auxiliary regularizer, encouraging attention that aligns with ground-truth discriminative regions, while suppressing attention elsewhere. We evaluated CAMAL across two learning paradigms -- Deep Learning (DL) and Deep Reinforcement Learning (DRL) -- and observed consistent, significant improvements in both attention alignment and faithfulness. In particular, CAMAL yields statistically significant gains in attention alignment across all settings, and improves attention faithfulness by over 35% compared to recent work. Moreover, we show that improved attention alignment and faithfulness enhance explainability, while yielding improved or comparable generalization performance without increasing inference cost. These findings demonstrate that the spatial information contained within segmentation masks can be effectively leveraged to guide model attention across learning tasks.

Related papers

• Focusing by Contrastive Attention: Enhancing VLMs' Visual Reasoning [79.34909830834464]
  Vision-Language Models (VLMs) have demonstrated remarkable success across diverse visual tasks, yet their performance degrades in complex visual environments.<n>We show that visual complexity strongly correlates with attention entropy, negatively impacting reasoning performance.<n>We propose Contrastive Attention Refinement for Visual Enhancement (CARVE), a training-free method that extracts task-relevant visual signals through attention contrasting at the pixel level.
  arXiv  Detail & Related papers  (2025-09-08T09:20:04Z)
• Mitigating Hallucination in Large Vision-Language Models via Adaptive Attention Calibration [2.19036693868242]
  Large vision-language models (LVLMs) achieve impressive performance on multimodal tasks but often suffer from hallucination.<n>We introduce the Confidence-Aware Attention (CAAC) framework to address this challenge by targeting two key biases.<n>CAAC employs a two-step approach: Visual-Token (VTC) to balance attention across visual tokens, and Adaptive Attention Re-Scaling (AAR) to reinforce visual grounding.
  arXiv  Detail & Related papers  (2025-05-27T17:45:21Z)
• Unveiling and Harnessing Hidden Attention Sinks: Enhancing Large Language Models without Training through Attention Calibration [15.36841874118801]
  We aim to provide a more profound understanding of the existence of attention sinks within large language models (LLMs) We propose a training-free Attention Technique (ACT) that automatically optimize the attention distributions on the fly during inference in an input-adaptive manner. ACT achieves an average improvement of up to 7.30% in accuracy across different datasets when applied to Llama-30B.
  arXiv  Detail & Related papers  (2024-06-22T07:00:43Z)
• Elliptical Attention [1.7597562616011944]
  Pairwise dot-product self-attention is key to the success of transformers that achieve state-of-the-art performance across a variety of applications in language and vision. We propose using a Mahalanobis distance metric for computing the attention weights to stretch the underlying feature space in directions of high contextual relevance.
  arXiv  Detail & Related papers  (2024-06-19T18:38:11Z)
• AlignZeg: Mitigating Objective Misalignment for Zero-shot Semantic Segmentation [123.88875931128342]
  A serious issue that harms the performance of zero-shot visual recognition is named objective misalignment. We propose a novel architecture named AlignZeg, which embodies a comprehensive improvement of the segmentation pipeline. Experiments demonstrate that AlignZeg markedly enhances zero-shot semantic segmentation.
  arXiv  Detail & Related papers  (2024-04-08T16:51:33Z)
• Dual Cross-Attention Learning for Fine-Grained Visual Categorization and Object Re-Identification [19.957957963417414]
  We propose a dual cross-attention learning (DCAL) algorithm to coordinate with self-attention learning. First, we propose global-local cross-attention (GLCA) to enhance the interactions between global images and local high-response regions. Second, we propose pair-wise cross-attention (PWCA) to establish the interactions between image pairs.
  arXiv  Detail & Related papers  (2022-05-04T16:14:26Z)
• Alignment Attention by Matching Key and Query Distributions [48.93793773929006]
  This paper introduces alignment attention that explicitly encourages self-attention to match the distributions of the key and query within each head. It is simple to convert any models with self-attention, including pre-trained ones, to the proposed alignment attention. On a variety of language understanding tasks, we show the effectiveness of our method in accuracy, uncertainty estimation, generalization across domains, and robustness to adversarial attacks.
  arXiv  Detail & Related papers  (2021-10-25T00:54:57Z)
• Counterfactual Attention Learning for Fine-Grained Visual Categorization and Re-identification [101.49122450005869]
  We present a counterfactual attention learning method to learn more effective attention based on causal inference. Specifically, we analyze the effect of the learned visual attention on network prediction. We evaluate our method on a wide range of fine-grained recognition tasks.
  arXiv  Detail & Related papers  (2021-08-19T14:53:40Z)
• More Than Just Attention: Learning Cross-Modal Attentions with Contrastive Constraints [63.08768589044052]
  We propose Contrastive Content Re-sourcing ( CCR) and Contrastive Content Swapping ( CCS) constraints to address such limitation. CCR and CCS constraints supervise the training of attention models in a contrastive learning manner without requiring explicit attention annotations. Experiments on both Flickr30k and MS-COCO datasets demonstrate that integrating these attention constraints into two state-of-the-art attention-based models improves the model performance.
  arXiv  Detail & Related papers  (2021-05-20T08:48:10Z)
• Heterogeneous Contrastive Learning: Encoding Spatial Information for Compact Visual Representations [183.03278932562438]
  This paper presents an effective approach that adds spatial information to the encoding stage to alleviate the learning inconsistency between the contrastive objective and strong data augmentation operations. We show that our approach achieves higher efficiency in visual representations and thus delivers a key message to inspire the future research of self-supervised visual representation learning.
  arXiv  Detail & Related papers  (2020-11-19T16:26:25Z)

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.