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 3 / 8 页

  1. NeurIPS 2026ShadowFPT: Backdooring Federated Prompt Tuning via Shadow Triggers
  2. NeurIPS 2026Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks
  3. NeurIPS 2026The Platonic Defense: Backdoor Defense for Self-Supervised Encoders in the Era of Large Scale Pre-training
  4. NeurIPS 2026Token by Token, Compromised: Backdoor Vulnerabilities in Unified Autoregressive Models
  5. NeurIPS 2026Towards Universal Black-box Attacks on Graph Neural Networks
  6. NeurIPS 2026TraceGuard: Defending Multi-Turn Jailbreak Attacks via Prompt-Response Risk Signal Tracking
  7. NeurIPS 2026Training-Based Backdoors Are Not Cryptographic
  8. NeurIPS 2026VOID: Backdoor Injection through Knowledge Vacuity in Federated Unlearning
  9. NeurIPS 2026When Sanitization Becomes the Trigger: Defense-Triggered Backdoor Attacks
  10. NeurIPS 2026Your Neighbors Know: Leveraging Local Neighborhoods for Backdoor Detection in Decentralized Learning
  11. CVPR 2025BadToken: Token-level Backdoor Attacks to Multi-modal Large Language Models
  12. CVPR 2025Black-Box Forgery Attacks on Semantic Watermarks for Diffusion Models
  13. CVPR 2025Can't Slow Me Down: Learning Robust and Hardware-Adaptive Object Detectors against Latency Attacks for Edge Devices
  14. CVPR 2025DeDe: Detecting Backdoor Samples for SSL Encoders via Decoders
  15. CVPR 2025Detecting Backdoor Attacks in Federated Learning via Direction Alignment Inspection
  16. CVPR 2025Deterministic Certification of Graph Neural Networks against Graph Poisoning Attacks with Arbitrary Perturbations
  17. CVPR 2025EntropyMark: Towards More Harmless Backdoor Watermark via Entropy-based Constraint for Open-source Dataset Copyright Protection
  18. CVPR 2025Improving Transferable Targeted Attacks with Feature Tuning Mixup
  19. CVPR 2025Infighting in the Dark: Multi-Label Backdoor Attack in Federated Learning
  20. CVPR 2025Invisible Backdoor Attack against Self-supervised Learning
  21. CVPR 2025Mind the Gap: Detecting Black-box Adversarial Attacks in the Making through Query Update Analysis
  22. CVPR 2025Model Poisoning Attacks to Federated Learning via Multi-Round Consistency
  23. CVPR 2025PSBD: Prediction Shift Uncertainty Unlocks Backdoor Detection
  24. CVPR 2025Revisiting Backdoor Attacks against Large Vision-Language Models from Domain Shift
  25. CVPR 2025Stealthy Backdoor Attack in Self-Supervised Learning Vision Encoders for Large Vision Language Models
  26. CVPR 2025Test-Time Backdoor Detection for Object Detection Models
  27. CVPR 2025UIBDiffusion: Universal Imperceptible Backdoor Attack for Diffusion Models
  28. CVPR 2025Variance-Based Membership Inference Attacks Against Large-Scale Image Captioning Models
  29. CVPR 2025Your Scale Factors are My Weapon: Targeted Bit-Flip Attacks on Vision Transformers via Scale Factor Manipulation
  30. ICLR 2025Activation Gradient based Poisoned Sample Detection Against Backdoor Attacks
  31. ICLR 2025Adversarial Training for Defense Against Label Poisoning Attacks
  32. ICLR 2025Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents
  33. ICLR 2025AI2TALE: An Innovative Information Theory-based Approach for Learning to Localize Phishing Attacks
  34. ICLR 2025Bad-PFL: Exploiting Backdoor Attacks against Personalized Federated Learning
  35. ICLR 2025Beyond Mere Token Analysis: A Hypergraph Metric Space Framework for Defending Against Socially Engineered LLM Attacks
  36. ICLR 2025Can We Trust Embodied Agents? Exploring Backdoor Attacks against Embodied LLM-Based Decision-Making Systems
  37. ICLR 2025Certifying Language Model Robustness with Fuzzed Randomized Smoothing: An Efficient Defense Against Backdoor Attacks
  38. ICLR 2025Concept-ROT: Poisoning Concepts in Large Language Models with Model Editing
  39. ICLR 2025Detecting Backdoor Samples in Contrastive Language Image Pretraining
  40. ICLR 2025DocMIA: Document-Level Membership Inference Attacks against DocVQA Models
  41. ICLR 2025Energy-based Backdoor Defense Against Federated Graph Learning
  42. ICLR 2025GPromptShield: Elevating Resilience in Graph Prompt Tuning Against Adversarial Attacks
  43. ICLR 2025Humanizing the Machine: Proxy Attacks to Mislead LLM Detectors
  44. ICLR 2025Injecting Universal Jailbreak Backdoors into LLMs in Minutes
  45. ICLR 2025Machine Unlearning Fails to Remove Data Poisoning Attacks
  46. ICLR 2025Mind Control through Causal Inference: Predicting Clean Images from Poisoned Data
  47. ICLR 2025Probe before You Talk: Towards Black-box Defense against Backdoor Unalignment for Large Language Models
  48. ICLR 2025Protecting against simultaneous data poisoning attacks
  49. ICLR 2025Provably Robust Explainable Graph Neural Networks against Graph Perturbation Attacks
  50. ICLR 2025REFINE: Inversion-Free Backdoor Defense via Model Reprogramming

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