后门攻击论文趋势 2023–2026

触发器与检测

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 381 篇被归入该主题的论文。

主要结论

每年论文数

各会议论文数

按年份和会议统计

后门攻击论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
20232115112673
202436272418105
20253125171992
202636203619111

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

全部后门攻击论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 380 篇，第 7 / 8 页

  1. NeurIPS 2024Trap-MID: Trapdoor-based Defense against Model Inversion Attacks
  2. NeurIPS 2024Tree of Attacks: Jailbreaking Black-Box LLMs Automatically
  3. NeurIPS 2024Uncovering, Explaining, and Mitigating the Superficial Safety of Backdoor Defense
  4. NeurIPS 2024Unelicitable Backdoors via Cryptographic Transformer Circuits
  5. NeurIPS 2024Unveiling and Mitigating Backdoor Vulnerabilities based on Unlearning Weight Changes and Backdoor Activeness
  6. NeurIPS 2024Watch Out for Your Agents! Investigating Backdoor Threats to LLM-Based Agents
  7. NeurIPS 2024WaveAttack: Asymmetric Frequency Obfuscation-based Backdoor Attacks Against Deep Neural Networks
  8. CVPR 2023Architectural Backdoors in Neural Networks
  9. CVPR 2023Backdoor Attacks Against Deep Image Compression via Adaptive Frequency Trigger
  10. CVPR 2023Backdoor Cleansing With Unlabeled Data
  11. CVPR 2023Backdoor Defense via Adaptively Splitting Poisoned Dataset
  12. CVPR 2023Backdoor Defense via Deconfounded Representation Learning
  13. CVPR 2023Can't Steal? Cont-Steal! Contrastive Stealing Attacks Against Image Encoders
  14. CVPR 2023Color Backdoor: A Robust Poisoning Attack in Color Space
  15. CVPR 2023Defending Against Patch-Based Backdoor Attacks on Self-Supervised Learning
  16. CVPR 2023Detecting Backdoors During the Inference Stage Based on Corruption Robustness Consistency
  17. CVPR 2023Detecting Backdoors in Pre-Trained Encoders
  18. CVPR 2023Dynamic Generative Targeted Attacks With Pattern Injection
  19. CVPR 2023Efficient Loss Function by Minimizing the Detrimental Effect of Floating-Point Errors on Gradient-Based Attacks
  20. CVPR 2023Enhancing the Self-Universality for Transferable Targeted Attacks
  21. CVPR 2023Ensemble-Based Blackbox Attacks on Dense Prediction
  22. CVPR 2023How to Backdoor Diffusion Models?
  23. CVPR 2023MEDIC: Remove Model Backdoors via Importance Driven Cloning
  24. CVPR 2023Minimizing Maximum Model Discrepancy for Transferable Black-Box Targeted Attacks
  25. CVPR 2023Progressive Backdoor Erasing via Connecting Backdoor and Adversarial Attacks
  26. CVPR 2023Re-Thinking Model Inversion Attacks Against Deep Neural Networks
  27. CVPR 2023Reinforcement Learning-Based Black-Box Model Inversion Attacks
  28. CVPR 2023Sibling-Attack: Rethinking Transferable Adversarial Attacks Against Face Recognition
  29. CVPR 2023Single Image Backdoor Inversion via Robust Smoothed Classifiers
  30. CVPR 2023The Best Defense Is a Good Offense: Adversarial Augmentation Against Adversarial Attacks
  31. CVPR 2023The Dark Side of Dynamic Routing Neural Networks: Towards Efficiency Backdoor Injection
  32. CVPR 2023TrojDiff: Trojan Attacks on Diffusion Models With Diverse Targets
  33. CVPR 2023You Are Catching My Attention: Are Vision Transformers Bad Learners Under Backdoor Attacks?
  34. ICLR 2023Adversarial Attacks on Adversarial Bandits
  35. ICLR 2023Certified Defences Against Adversarial Patch Attacks on Semantic Segmentation
  36. ICLR 2023Clean-image Backdoor: Attacking Multi-label Models with Poisoned Labels Only
  37. ICLR 2023Distilling Cognitive Backdoor Patterns within an Image
  38. ICLR 2023Effective passive membership inference attacks in federated learning against overparameterized models
  39. ICLR 2023Few-shot Backdoor Attacks via Neural Tangent Kernels
  40. ICLR 2023FLIP: A Provable Defense Framework for Backdoor Mitigation in Federated Learning
  41. ICLR 2023Incompatibility Clustering as a Defense Against Backdoor Poisoning Attacks
  42. ICLR 2023Indiscriminate Poisoning Attacks on Unsupervised Contrastive Learning
  43. ICLR 2023Revisiting the Assumption of Latent Separability for Backdoor Defenses
  44. ICLR 2023SCALE-UP: An Efficient Black-box Input-level Backdoor Detection via Analyzing Scaled Prediction Consistency
  45. ICLR 2023The Dark Side of AutoML: Towards Architectural Backdoor Search
  46. ICLR 2023Towards Robustness Certification Against Universal Perturbations
  47. ICLR 2023TrojText: Test-time Invisible Textual Trojan Insertion
  48. ICLR 2023UNICORN: A Unified Backdoor Trigger Inversion Framework
  49. ICML 2023Are Diffusion Models Vulnerable to Membership Inference Attacks?
  50. ICML 2023Data Poisoning Attacks Against Multimodal Encoders

在 Claude、Cursor 里直接问这个趋势

接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。

可以直接复制这句话问 AI

用 paperlist 查后门攻击在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？

AI 会调用的工具：get_topic_stats {"topic": "backdoor-attacks"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

常见问题

2023–2026 年四大会议一共有多少篇后门攻击论文？
共 381 篇，按年份分别是 2023 年 73 篇、2024 年 105 篇、2025 年 92 篇、2026 年 111 篇。
NeurIPS 2026 有多少篇后门攻击论文？
NeurIPS 2026 有 36 篇后门攻击论文，上一届是 31 篇。
ICLR 2026 有多少篇后门攻击论文？
ICLR 2026 有 20 篇后门攻击论文，上一届是 25 篇。
ICML 2026 有多少篇后门攻击论文？
ICML 2026 有 36 篇后门攻击论文，上一届是 17 篇。
CVPR 2026 有多少篇后门攻击论文？
CVPR 2026 有 19 篇后门攻击论文，上一届是 19 篇。
哪个会议的后门攻击论文最多？
NeurIPS，2023–2026 年共 124 篇。

「安全与可信」下的其他主题