--- title: NDSS 2025 – Defending Against Backdoor Attacks On Graph... description: DShield is a defense framework presented at NDSS 2025 that protects Graph Neural Networks (GNNs) from backdoor attacks using a discrepancy learning mechanism.... canonical: https://daily.dev/posts/ndss-2025-defending-against-backdoor-attacks-on-graph-neural-networks-via-discrepancy-learning-6swnofjpb twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: NDSS 2025 – Defending Against Backdoor Attacks On Graph Neural Networks Via Discrepancy Learning | daily.dev og:description: DShield is a defense framework presented at NDSS 2025 that protects Graph Neural Networks (GNNs) from backdoor attacks using a discrepancy learning mechanism.... og:url: https://daily.dev/posts/ndss-2025-defending-against-backdoor-attacks-on-graph-neural-networks-via-discrepancy-learning-6swnofjpb og:image: https://api.daily.dev/og/posts/6sWnOfJpB.png og:image:alt: NDSS 2025 – Defending Against Backdoor Attacks On Graph Neural Networks Via Discrepancy Learning og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # NDSS 2025 – Defending Against Backdoor Attacks On Graph Neural Networks Via Discrepancy Learning **[Security Boulevard](https://daily.dev/sources/securityboulevard)** · 1 min read · 0 upvotes · 0 comments ## Summary DShield is a defense framework presented at NDSS 2025 that protects Graph Neural Networks (GNNs) from backdoor attacks using a discrepancy learning mechanism. It addresses two key attack behaviors: semantic drift in dirty-label attacks and attribute over-emphasis in clean-label attacks. By combining a self-supervised model with a backdoored model, DShield identifies and filters poisoned nodes without relying on manipulated labels. Evaluated against 21 backdoor attacks across 7 datasets, DShield reduces attack success rates dramatically—on the Cora dataset, from 54.47% (second-best defense) down to 1.33%—while maintaining 82.15% accuracy on clean nodes. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Similar posts on daily.dev - [NDSS 2025 – BitShield: Defending Against Bit-Flip Attacks On DNN Executables](https://daily.dev/posts/ndss-2025-bitshield-defending-against-bit-flip-attacks-on-dnn-executables-8fpa84bsd) · Security Boulevard · 0 upvotes · 0 comments - [NDSS 2025 – Revisiting Concept Drift In Windows Malware Detection](https://daily.dev/posts/ndss-2025-revisiting-concept-drift-in-windows-malware-detection-esqdwn7iy) · Security Boulevard · 1 upvotes · 0 comments - [NDSS 2025 – DLBox: New Model Training Framework For Protecting Training Data](https://daily.dev/posts/ndss-2025-dlbox-new-model-training-framework-for-protecting-training-data-7bvtk9tnh) · Security Boulevard · 0 upvotes · 0 comments - [DDN: Discrete Distribution Networks](https://daily.dev/posts/ddn-discrete-distribution-networks-m5tpdt8nd) · Hacker News · 0 upvotes · 0 comments - [Architectural awareness: building the sensor into the boat](https://daily.dev/posts/architectural-awareness-building-the-sensor-into-the-boat-3b4zwhyth) · Medium · 0 upvotes · 0 comments --- Tags: [#deep-learning](https://daily.dev/tags/deep-learning) [View this post on daily.dev](https://daily.dev/posts/ndss-2025-defending-against-backdoor-attacks-on-graph-neural-networks-via-discrepancy-learning-6swnofjpb) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"NDSS 2025 – Defending Against Backdoor Attacks On Graph Neural Networks Via Discrepancy Learning","url":"https://daily.dev/posts/ndss-2025-defending-against-backdoor-attacks-on-graph-neural-networks-via-discrepancy-learning-6swnofjpb","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/ndss-2025-defending-against-backdoor-attacks-on-graph-neural-networks-via-discrepancy-learning-6swnofjpb"},"datePublished":"2026-02-19T16:34:00.772Z","dateModified":"2026-02-19T16:34:22.600Z","description":"DShield is a defense framework presented at NDSS 2025 that protects Graph Neural Networks (GNNs) from backdoor attacks using a discrepancy learning mechanism....","image":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/8fe9eeb967de205fd6743e2535f16178?_a=AQAEuop","thumbnailUrl":"https://media.daily.dev/image/upload/f_auto,q_auto/v1/posts/8fe9eeb967de205fd6743e2535f16178?_a=AQAEuop","isAccessibleForFree":true,"articleSection":"Security Boulevard","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Security Boulevard","logo":"https://media.daily.dev/image/upload/t_logo,f_auto/v1/logos/3613c832180040de8d85bb29f74395be","url":"https://daily.dev/sources/securityboulevard"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/ndss-2025-defending-against-backdoor-attacks-on-graph-neural-networks-via-discrepancy-learning-6swnofjpb","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"deep-learning","timeRequired":"PT1M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Security Boulevard","item":"https://daily.dev/sources/securityboulevard"},{"@type":"ListItem","position":3,"name":"NDSS 2025 – Defending Against Backdoor Attacks On Graph Neural Networks Via Discrepancy Learning"}]} ```