Related papers: Forewarned is Forearmed: When Non-Sequential Embedding Turns Into an Anomaly Detector

Forewarned is Forearmed: When Non-Sequential Embedding Turns Into an Anomaly Detector

URL: http://arxiv.org/abs/2606.30196v1
Date: Mon, 29 Jun 2026 12:12:43 GMT
Title: Forewarned is Forearmed: When Non-Sequential Embedding Turns Into an Anomaly Detector
Abstract summary: We show that certain embedding dimensions are sensitive to perturbations and can serve as indicators of decoding anomalies.<n>This work underscores the importance of understanding and analyzing the embeddings themselves to enhance the reliability of multimodal representations.
Score: 0.7095058159492491
License: http://creativecommons.org/licenses/by-nc-sa/4.0/
Abstract: This paper offers an in-depth analysis of non-sequential multimodal sentence-level embeddings, with a particular focus on the SONAR model. We demonstrate that certain embedding dimensions are sensitive to perturbations and can serve as indicators of decoding anomalies. By leveraging the consistency between successive encoding and decoding, we successfully build an accurate detector. Additionally, we explore modifying specific dimensions of interest to attempt to correct them. This work underscores the importance of understanding and analyzing the embeddings themselves to enhance the reliability of multimodal representations.

Related papers

An Encode-then-Decompose Approach to Unsupervised Time Series Anomaly Detection on Contaminated Training Data--Extended Version [31.029023011951793]
Time series anomaly detection is important in modern large-scale systems to analyze and monitor the operation of diverse systems.<n>Unsupervised approaches have received widespread interest, as they do not require anomaly labels during training.<n>We propose a novel encode-then-decompose paradigm, where we decompose the encoded representation into stable and auxiliary representations.
arXiv Detail & Related papers (2025-10-21T18:23:20Z)
Distributional Information Embedding: A Framework for Multi-bit Watermarking [35.319577498993354]
This paper introduces a novel problem, distributional information embedding, motivated by the practical demands of watermarking for large language models (LLMs)<n>We develop an information-theoretic framework to analyze this distributional information embedding problem.<n>We identify schemes that maximize detection probability while adhering to constraints on false alarm and distortion.
arXiv Detail & Related papers (2025-01-27T23:01:56Z)
Mitigating Shortcut Learning with Diffusion Counterfactuals and Diverse Ensembles [104.60508550106618]
We propose DiffDiv, an ensemble diversification framework exploiting Diffusion Probabilistic Models (DPMs)<n>We show that DPMs can generate images with novel feature combinations, even when trained on samples displaying correlated input features.<n>We show that DPM-guided diversification is sufficient to remove dependence on shortcut cues, without a need for additional supervised signals.
arXiv Detail & Related papers (2023-11-23T15:47:33Z)
Transcending Forgery Specificity with Latent Space Augmentation for Generalizable Deepfake Detection [57.646582245834324]
We propose a simple yet effective deepfake detector called LSDA. It is based on a idea: representations with a wider variety of forgeries should be able to learn a more generalizable decision boundary. We show that our proposed method is surprisingly effective and transcends state-of-the-art detectors across several widely used benchmarks.
arXiv Detail & Related papers (2023-11-19T09:41:10Z)
Disentanglement via Latent Quantization [60.37109712033694]
In this work, we construct an inductive bias towards encoding to and decoding from an organized latent space. We demonstrate the broad applicability of this approach by adding it to both basic data-re (vanilla autoencoder) and latent-reconstructing (InfoGAN) generative models.
arXiv Detail & Related papers (2023-05-28T06:30:29Z)
The Treasure Beneath Multiple Annotations: An Uncertainty-aware Edge Detector [70.43599299422813]
Existing methods fuse multiple annotations using a simple voting process, ignoring the inherent ambiguity of edges and labeling bias of annotators. We propose a novel uncertainty-aware edge detector (UAED), which employs uncertainty to investigate the subjectivity and ambiguity of diverse annotations. UAED achieves superior performance consistently across multiple edge detection benchmarks.
arXiv Detail & Related papers (2023-03-21T13:14:36Z)
Uncertainty Quantification in Medical Image Segmentation with Multi-decoder U-Net [3.961279440272763]
We exploit the medical image segmentation uncertainty by measuring segmentation performance with multiple annotations in a supervised learning manner. We propose a U-Net based architecture with multiple decoders, where the image representation is encoded with the same encoder, and segmentation referring to each annotation is estimated with multiple decoders. The proposed architecture is trained in an end-to-end manner and able to improve predictive uncertainty estimates.
arXiv Detail & Related papers (2021-09-15T01:46:29Z)
Bayesian Attention Belief Networks [59.183311769616466]
Attention-based neural networks have achieved state-of-the-art results on a wide range of tasks. This paper introduces Bayesian attention belief networks, which construct a decoder network by modeling unnormalized attention weights. We show that our method outperforms deterministic attention and state-of-the-art attention in accuracy, uncertainty estimation, generalization across domains, and adversarial attacks.
arXiv Detail & Related papers (2021-06-09T17:46:22Z)
ESAD: End-to-end Deep Semi-supervised Anomaly Detection [85.81138474858197]
We propose a new objective function that measures the KL-divergence between normal and anomalous data. The proposed method significantly outperforms several state-of-the-arts on multiple benchmark datasets.
arXiv Detail & Related papers (2020-12-09T08:16:35Z)

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