Detection Transformers Under the Knife: A Neuroscience-Inspired Approach to Ablations

• URL: http://arxiv.org/abs/2507.21723v1
• Date: Tue, 29 Jul 2025 12:00:08 GMT
• Title: Detection Transformers Under the Knife: A Neuroscience-Inspired Approach to Ablations
• Abstract summary: We systematically analyze the impact of ablating key components in three state-of-the-art detection transformer models.<n>We evaluate the effects of these ablations on the performance metrics gIoU and F1-score.<n>This study advances XAI for DETRs by clarifying the contributions of internal components to model performance.
• Score: 5.5967570276373655
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In recent years, Explainable AI has gained traction as an approach to enhancing model interpretability and transparency, particularly in complex models such as detection transformers. Despite rapid advancements, a substantial research gap remains in understanding the distinct roles of internal components - knowledge that is essential for improving transparency and efficiency. Inspired by neuroscientific ablation studies, which investigate the functions of brain regions through selective impairment, we systematically analyze the impact of ablating key components in three state-of-the-art detection transformer models: Detection transformer (DETR), deformable detection transformer (DDETR), and DETR with improved denoising anchor boxes (DINO). The ablations target query embeddings, encoder and decoder multi-head self-attentions (MHSA) as well as decoder multi-head cross-attention (MHCA) layers. We evaluate the effects of these ablations on the performance metrics gIoU and F1-score, quantifying effects on both the classification and regression sub-tasks on the COCO dataset. To facilitate reproducibility and future research, we publicly release the DeepDissect library. Our findings reveal model-specific resilience patterns: while DETR is particularly sensitive to ablations in encoder MHSA and decoder MHCA, DDETR's multi-scale deformable attention enhances robustness, and DINO exhibits the greatest resilience due to its look-forward twice update rule, which helps distributing knowledge across blocks. These insights also expose structural redundancies, particularly in DDETR's and DINO's decoder MHCA layers, highlighting opportunities for model simplification without sacrificing performance. This study advances XAI for DETRs by clarifying the contributions of internal components to model performance, offering insights to optimize and improve transparency and efficiency in critical applications.

Related papers

• Large Vision-Language Models Get Lost in Attention [51.851592109135716]
  We propose a unified framework grounded in information theory and geometry to quantify the geometric and entropic nature of residual updates.<n>Applying this unified framework reveals a fundamental functional decoupling: Attention acts as a subspace-language operator focused on reconfiguration, whereas FFNs serve as subspace-language operators driving semantic innovation.
  arXiv  Detail & Related papers  (2026-05-07T04:45:52Z)
• Neural Dynamics Self-Attention for Spiking Transformers [59.57278823240336]
  Spiking Neural Networks (SNNs) with Transformer architectures offer a promising pathway to balance energy efficiency and performance.<n>Existing Spiking Transformers face two critical challenges: (i) a substantial performance gap compared to their Artificial Neural Networks (ANNs) counterparts and (ii) high memory overhead during inference.<n>We propose LRF-Dyn, which uses spiking neurons with localized receptive fields to compute attention while reducing memory requirements.
  arXiv  Detail & Related papers  (2026-03-09T12:23:37Z)
• Multi-Agent Collaborative Intrusion Detection for Low-Altitude Economy IoT: An LLM-Enhanced Agentic AI Framework [60.72591149679355]
  The rapid expansion of low-altitude economy Internet of Things (LAE-IoT) networks has created unprecedented security challenges.<n>Traditional intrusion detection systems fail to tackle the unique characteristics of aerial IoT environments.<n>We introduce a large language model (LLM)-enabled agentic AI framework for enhancing intrusion detection in LAE-IoT networks.
  arXiv  Detail & Related papers  (2026-01-25T12:47:25Z)
• EEG-D3: A Solution to the Hidden Overfitting Problem of Deep Learning Models [33.22173262655432]
  We introduce Disentangled Decoding Decomposition (D3), a weakly supervised method for training deep learning models across EEG datasets.<n>By predicting the place in the respective trial sequence from which the input window was sampled, EEG-D3 separates latent components of brain activity.<n>We exploit the linearly separable latent space for effective few-shot learning on sleep stage classification.
  arXiv  Detail & Related papers  (2025-12-15T19:00:10Z)
• FAIM: Frequency-Aware Interactive Mamba for Time Series Classification [87.84511960413715]
  Time series classification (TSC) is crucial in numerous real-world applications, such as environmental monitoring, medical diagnosis, and posture recognition.<n>We propose FAIM, a lightweight Frequency-Aware Interactive Mamba model.<n>We show that FAIM consistently outperforms existing state-of-the-art (SOTA) methods, achieving a superior trade-off between accuracy and efficiency.
  arXiv  Detail & Related papers  (2025-11-26T08:36:33Z)
• Improving Deepfake Detection with Reinforcement Learning-Based Adaptive Data Augmentation [60.04281435591454]
  CRDA (Curriculum Reinforcement-Learning Data Augmentation) is a novel framework guiding detectors to progressively master multi-domain forgery features.<n>Central to our approach is integrating reinforcement learning and causal inference.<n>Our method significantly improves detector generalizability, outperforming SOTA methods across multiple cross-domain datasets.
  arXiv  Detail & Related papers  (2025-11-10T12:45:52Z)
• Source-Free Object Detection with Detection Transformer [59.33653163035064]
  Source-Free Object Detection (SFOD) enables knowledge transfer from a source domain to an unsupervised target domain for object detection without access to source data.<n>Most existing SFOD approaches are either confined to conventional object detection (OD) models like Faster R-CNN or designed as general solutions without tailored adaptations for novel OD architectures, especially Detection Transformer (DETR)<n>In this paper, we introduce Feature Reweighting ANd Contrastive Learning NetworK (FRANCK), a novel SFOD framework specifically designed to perform query-centric feature enhancement for DETRs.
  arXiv  Detail & Related papers  (2025-10-13T07:35:04Z)
• Transformer-Based Indirect Structural Health Monitoring of Rail Infrastructure with Attention-Driven Detection and Localization of Transient Defects [1.1782896991259]
  We introduce an incremental synthetic data benchmark designed to evaluate model robustness against progressively complex challenges.<n>We evaluate several established unsupervised models alongside our proposed Attention-Focused Transformer.<n>Our proposed model achieves accuracy comparable to the state-of-the-art solution while demonstrating better inference speed.
  arXiv  Detail & Related papers  (2025-10-08T23:01:53Z)
• CEReBrO: Compact Encoder for Representations of Brain Oscillations Using Efficient Alternating Attention [53.539020807256904]
  We introduce a Compact for Representations of Brain Oscillations using alternating attention (CEReBrO)<n>Our tokenization scheme represents EEG signals at a per-channel patch.<n>We propose an alternating attention mechanism that jointly models intra-channel temporal dynamics and inter-channel spatial correlations, achieving 2x speed improvement with 6x less memory required compared to standard self-attention.
  arXiv  Detail & Related papers  (2025-01-18T21:44:38Z)
• Localized Gaussians as Self-Attention Weights for Point Clouds Correspondence [92.07601770031236]
  We investigate semantically meaningful patterns in the attention heads of an encoder-only Transformer architecture. We find that fixing the attention weights not only accelerates the training process but also enhances the stability of the optimization.
  arXiv  Detail & Related papers  (2024-09-20T07:41:47Z)
• CycleHOI: Improving Human-Object Interaction Detection with Cycle Consistency of Detection and Generation [37.45945633515955]
  We propose a new learning framework, coined as CycleHOI, to boost the performance of human-object interaction (HOI) detection. Our key design is to introduce a novel cycle consistency loss for the training of HOI detector. We perform extensive experiments to verify the effectiveness and generalization power of our CycleHOI.
  arXiv  Detail & Related papers  (2024-07-16T06:55:43Z)
• An Attention-Based Deep Generative Model for Anomaly Detection in Industrial Control Systems [3.303448701376485]
  Anomaly detection is critical for the secure and reliable operation of industrial control systems. This paper presents a novel deep generative model to meet this need.
  arXiv  Detail & Related papers  (2024-05-03T23:58:27Z)
• DetDiffusion: Synergizing Generative and Perceptive Models for Enhanced Data Generation and Perception [78.26734070960886]
  Current perceptive models heavily depend on resource-intensive datasets. We introduce perception-aware loss (P.A. loss) through segmentation, improving both quality and controllability. Our method customizes data augmentation by extracting and utilizing perception-aware attribute (P.A. Attr) during generation.
  arXiv  Detail & Related papers  (2024-03-20T04:58:03Z)
• OCR is All you need: Importing Multi-Modality into Image-based Defect Detection System [7.1083241462091165]
  We introduce an external modality-guided data mining framework, primarily rooted in optical character recognition (OCR), to extract statistical features from images. A key aspect of our approach is the alignment of external modality features, extracted using a single modality-aware model, with image features encoded by a convolutional neural network. Our methodology considerably boosts the recall rate of the defect detection model and maintains high robustness even in challenging scenarios.
  arXiv  Detail & Related papers  (2024-03-18T07:41:39Z)
• InfoRM: Mitigating Reward Hacking in RLHF via Information-Theoretic Reward Modeling [66.3072381478251]
  Reward hacking, also termed reward overoptimization, remains a critical challenge. We propose a framework for reward modeling, namely InfoRM, by introducing a variational information bottleneck objective. We show that InfoRM's overoptimization detection mechanism is not only effective but also robust across a broad range of datasets.
  arXiv  Detail & Related papers  (2024-02-14T17:49:07Z)
• BDHT: Generative AI Enables Causality Analysis for Mild Cognitive Impairment [34.60961915466469]
  A brain diffuser with hierarchical transformer (BDHT) is proposed to estimate effective connectivity for mild cognitive impairment (MCI) analysis. The proposed model achieves superior performance in terms of accuracy and robustness compared to existing approaches.
  arXiv  Detail & Related papers  (2023-12-14T15:12:00Z)
• ADT: Agent-based Dynamic Thresholding for Anomaly Detection [4.356615197661274]
  We propose an agent-based dynamic thresholding (ADT) framework based on a deep Q-network. An auto-encoder is utilized in this study to obtain feature representations and produce anomaly scores for complex input data. ADT can adjust thresholds adaptively by utilizing the anomaly scores from the auto-encoder.
  arXiv  Detail & Related papers  (2023-12-03T19:07:30Z)
• A Generic Shared Attention Mechanism for Various Backbone Neural Networks [53.36677373145012]
  Self-attention modules (SAMs) produce strongly correlated attention maps across different layers. Dense-and-Implicit Attention (DIA) shares SAMs across layers and employs a long short-term memory module. Our simple yet effective DIA can consistently enhance various network backbones.
  arXiv  Detail & Related papers  (2022-10-27T13:24:08Z)
• Exploring Inconsistent Knowledge Distillation for Object Detection with Data Augmentation [66.25738680429463]
  Knowledge Distillation (KD) for object detection aims to train a compact detector by transferring knowledge from a teacher model. We propose inconsistent knowledge distillation (IKD) which aims to distill knowledge inherent in the teacher model's counter-intuitive perceptions. Our method outperforms state-of-the-art KD baselines on one-stage, two-stage and anchor-free object detectors.
  arXiv  Detail & Related papers  (2022-09-20T16:36:28Z)
• Is Disentanglement enough? On Latent Representations for Controllable Music Generation [78.8942067357231]
  In the absence of a strong generative decoder, disentanglement does not necessarily imply controllability. The structure of the latent space with respect to the VAE-decoder plays an important role in boosting the ability of a generative model to manipulate different attributes.
  arXiv  Detail & Related papers  (2021-08-01T18:37:43Z)
• DA-DETR: Domain Adaptive Detection Transformer with Information Fusion [53.25930448542148]
  DA-DETR is a domain adaptive object detection transformer that introduces information fusion for effective transfer from a labeled source domain to an unlabeled target domain. We introduce a novel CNN-Transformer Blender (CTBlender) that fuses the CNN features and Transformer features ingeniously for effective feature alignment and knowledge transfer across domains. CTBlender employs the Transformer features to modulate the CNN features across multiple scales where the high-level semantic information and the low-level spatial information are fused for accurate object identification and localization.
  arXiv  Detail & Related papers  (2021-03-31T13:55:56Z)
• Anomaly Detection Based on Selection and Weighting in Latent Space [73.01328671569759]
  We propose a novel selection-and-weighting-based anomaly detection framework called SWAD. Experiments on both benchmark and real-world datasets have shown the effectiveness and superiority of SWAD.
  arXiv  Detail & Related papers  (2021-03-08T10:56:38Z)

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.