mlsanityData to 14 Oct 2026

Task-Agnostic Detector for Insertion-Based Backdoor Attacks

arXiv 2403.17155 ↗First posted 25 Mar 2024cs.CL cs.CRcited by 23

Not on the map. Why?

Abstract (arXiv)

Textual backdoor attacks pose significant security threats. Current detection approaches, typically relying on intermediate feature representation or reconstructing potential triggers, are task-specific and less effective beyond sentence classification, struggling with tasks like question answering and named entity recognition. We introduce TABDet (Task-Agnostic Backdoor Detector), a pioneering task-agnostic method for backdoor detection. TABDet leverages final layer logits combined with an efficient pooling technique, enabling unified logit representation across three prominent NLP tasks. TABDet can jointly learn from diverse task-specific models, demonstrating superior detection efficacy over traditional task-specific methods.

Key facts. Cited by 23 papers in the map (data through 14 Oct 2026). Builds on 32 of its 68 references that are in the map. First posted on arXiv 25 Mar 2024.

Builds on

32 of 68 references.

References · 32

  1. 2017Attention Is All You Needcited by 60,361
  2. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,399
  3. 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,372
  4. 2019DistilBERT, a distilled version of BERT: smaller, faster, cheaper and lightercited by 4,307
  5. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,148
  6. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,106
  7. 2020Weight Poisoning Attacks on Pre-trained Modelscited by 341
  8. 2020ONION: A Simple and Effective Defense Against Textual Backdoor Attackscited by 263
  9. 2019A backdoor attack against LSTM-based text classification systemscited by 227
  10. 2021Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Triggercited by 222
  11. 2020BadNL: Backdoor Attacks against NLP Models with Semantic-preserving Improvementscited by 211
  12. 2021Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfercited by 158
20 more
  1. 2021Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of the Embedding Layers in NLP Modelscited by 105
  2. 2021RAP: Robustness-Aware Perturbations for Defending against Backdoor Attacks on NLP Modelscited by 104
  3. 2022A Unified Evaluation of Textual Backdoor Learning: Frameworks and Benchmarkscited by 74
  4. 2021T-Miner: A Generative Approach to Defend Against Trojan Attacks on DNN-based Text Classificationcited by 60
  5. 2022A Study of the Attention Abnormality in Trojaned BERTscited by 48
  6. 2022A Multimodal Transformer: Fusing Clinical Notes with Structured EHR Data for Interpretable In-Hospital Mortality Predictioncited by 48
  7. 2022Towards Robust Ranker for Text Retrievalcited by 37
  8. 2021ReLAX: Reinforcement Learning Agent eXplainer for Arbitrary Predictive Modelscited by 25
  9. 2022Defending Graph Convolutional Networks against Dynamic Graph Perturbations via Bayesian Self-supervisioncited by 16
  10. 2021TopoTxR: A Topological Biomarker for Predicting Treatment Response in Breast Cancercited by 13
  11. 2023BotTriNet: A Unified and Efficient Embedding for Social Bots Detection via Metric Learningcited by 13
  12. 2022Robust Node Classification on Graphs: Jointly from Bayesian Label Transition and Topology-based Label Propagationcited by 12
  13. 2021Distractor-Aware Fast Tracking via Dynamic Convolutions and MOT Philosophycited by 11
  14. 2021DeepMTL Pro: Deep Learning Based MultipleTransmitter Localization and Power Estimationcited by 10
  15. 2021OmniLytics: A Blockchain-based Secure Data Market for Decentralized Machine Learningcited by 9
  16. 2023FineEHR: Refine Clinical Note Representations to Improve Mortality Predictioncited by 9
  17. 2019Transfer Learning for Scientific Data Chain Extraction in Small Chemical Corpus with BERT-CRF Modelcited by 8
  18. 2024Comprehensive evaluation of Mal-API-2019 dataset by machine learning in malware detectioncited by 8
  19. 2023The Dark Side of Explanations: Poisoning Recommender Systems with Counterfactual Examplescited by 7
  20. 2023RIDE: Real-time Intrusion Detection via Explainable Machine Learning Implemented in a Memristor Hardware Architecturecited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 17; 2025: 4; 2026: 2.

23 papers.

Sort:

By area:

  1. 2024JailbreakZoo: Survey, Landscapes, and Horizons in Jailbreaking Large Language and Vision-Language Modelscited by 70
  2. 2024"Glue pizza and eat rocks" -- Exploiting Vulnerabilities in Retrieval-Augmented Generative Modelscited by 60
  3. 2024TrojVLM: Backdoor Attack Against Vision Language Modelscited by 39
  4. 2025textitAgents Under Siege: Breaking Pragmatic Multi-Agent LLM Systems with Optimized Prompt Attackscited by 34
  5. 2024Backdooring Vision-Language Models with Out-Of-Distribution Datacited by 33
  6. 2024Deconstructing The Ethics of Large Language Models from Long-standing Issues to New-emerging Dilemmas: A Surveycited by 23
  7. 2024MAPO: Boosting Large Language Model Performance with Model-Adaptive Prompt Optimizationcited by 21
  8. 2024InsectMamba: Insect Pest Classification with State Space Modelcited by 11
  9. 2024A Study of Backdoors in Instruction Fine-tuned Language Modelscited by 11
  10. 2024Research on Driver Facial Fatigue Detection Based on Yolov8 Modelcited by 11
  11. 2024Data Augmentation for Image Classification using Generative AIcited by 11
  12. 2024Hummer: Towards Limited Competitive Preference Datasetcited by 10
11 more
  1. 2024Optimizing Search Advertising Strategies: Integrating Reinforcement Learning with Generalized Second-Price Auctions for Enhanced Ad Ranking and Biddingcited by 9
  2. 2024Deep Learning in Medical Image Classification from MRI-based Brain Tumor Imagescited by 7
  3. 2025DREAM: Disentangling Risks to Enhance Safety Alignment in Multimodal Large Language Modelscited by 6
  4. 2024BadCLM: Backdoor Attack in Clinical Language Models for Electronic Health Recordscited by 5
  5. 2025Credit Risk Identification in Supply Chains Using Generative Adversarial Networkscited by 5
  6. 2024Federated Learning for Smart Grid: A Survey on Applications and Potential Vulnerabilitiescited by 3
  7. 2024A Survey on the Application of Generative Adversarial Networks in Cybersecurity: Prospective, Direction and Open Research Scopescited by 2
  8. 2024Analyzing Multi-Head Attention on Trojan BERT Modelscited by 0
  9. 2025Inverting Trojans in LLMscited by 0
  10. 2026CSO-LLM: Class Subspace Orthogonalization for Post-Training Backdoor Detection and Trigger Inversion in LLMscited by 0
  11. 2026Backdoor Learning in Language Models and Vision-Language Modelscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2021BadPre: Task-agnostic Backdoor Attacks to Pre-trained NLP Foundation Models
  2. 2023Prompt as Triggers for Backdoor Attack: Examining the Vulnerability in Language Models
  3. 2024SynGhost: Invisible and Universal Task-agnostic Backdoor Attack via Syntactic Transfer
  4. 2021Turn the Combination Lock: Learnable Textual Backdoor Attacks via Word Substitution
  5. 2022BITE: Textual Backdoor Attacks with Iterative Trigger Injection
  6. 2022Expose Backdoors on the Way: A Feature-Based Efficient Defense against Textual Backdoor Attacks
  7. 2022Detecting Backdoors in Deep Text Classifiers
  8. 2025Unmasking Backdoors: An Explainable Defense via Gradient-Attention Anomaly Scoring for Pre-trained Language Models