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 4 / 8 页

  1. ICLR 2025RobustKV: Defending Large Language Models against Jailbreak Attacks via KV Eviction
  2. ICLR 2025Robustness Inspired Graph Backdoor Defense
  3. ICLR 2025UV-Attack: Physical-World Adversarial Attacks on Person Detection via Dynamic-NeRF-based UV Mapping
  4. ICLR 2025Wicked Oddities: Selectively Poisoning for Effective Clean-Label Backdoor Attacks
  5. ICML 2025A Closer Look at Backdoor Attacks on CLIP
  6. ICML 2025Adversarial Inception Backdoor Attacks against Reinforcement Learning
  7. ICML 2025Backdoor Attacks in Token Selection of Attention Mechanism
  8. ICML 2025Bi-perspective Splitting Defense: Achieving Clean-Seed-Free Backdoor Security
  9. ICML 2025Circumventing Backdoor Space via Weight Symmetry
  10. ICML 2025CROW: Eliminating Backdoors from Large Language Models via Internal Consistency Regularization
  11. ICML 2025Does Low Rank Adaptation Lead to Lower Robustness against Training-Time Attacks?
  12. ICML 2025Emoji Attack: Enhancing Jailbreak Attacks Against Judge LLM Detection
  13. ICML 2025Flexible, Efficient, and Stable Adversarial Attacks on Machine Unlearning
  14. ICML 2025HyperNear: Unnoticeable Node Injection Attacks on Hypergraph Neural Networks
  15. ICML 2025ICLShield: Exploring and Mitigating In-Context Learning Backdoor Attacks
  16. ICML 2025Optimizing Adaptive Attacks against Watermarks for Language Models
  17. ICML 2025PoisonBench: Assessing Language Model Vulnerability to Poisoned Preference Data
  18. ICML 2025SPMC: Self-Purifying Federated Backdoor Defense via Margin Contribution
  19. ICML 2025Test-Time Multimodal Backdoor Detection by Contrastive Prompting
  20. ICML 2025The Ripple Effect: On Unforeseen Complications of Backdoor Attacks
  21. ICML 2025Trustworthy Machine Learning through Data-Specific Indistinguishability
  22. NeurIPS 2025A Set of Generalized Components to Achieve Effective Poison-only Clean-label Backdoor Attacks with Collaborative Sample Selection and Triggers
  23. NeurIPS 2025Adversarial Paraphrasing: A Universal Attack for Humanizing AI-Generated Text
  24. NeurIPS 2025Attack by Yourself: Effective and Unnoticeable Multi-Category Graph Backdoor Attacks with Subgraph Triggers Pool
  25. NeurIPS 2025Backdoor Mitigation via Invertible Pruning Masks
  26. NeurIPS 2025BackdoorDM: A Comprehensive Benchmark for Backdoor Learning on Diffusion Model
  27. NeurIPS 2025BackdoorLLM: A Comprehensive Benchmark for Backdoor Attacks and Defenses on Large Language Models
  28. NeurIPS 2025BadVLA: Towards Backdoor Attacks on Vision-Language-Action Models via Objective-Decoupled Optimization
  29. NeurIPS 2025Competitive Advantage Attacks to Decentralized Federated Learning
  30. NeurIPS 2025Defending Multimodal Backdoored Models by Repulsive Visual Prompt Tuning
  31. NeurIPS 2025Efficient Federated Learning against Byzantine Attacks and Data Heterogeneity via Aggregating Normalized Gradients
  32. NeurIPS 2025Enhancing LLM Watermark Resilience Against Both Scrubbing and Spoofing Attacks
  33. NeurIPS 2025FedRACE: A Hierarchical and Statistical Framework for Robust Federated Learning
  34. NeurIPS 2025From Dormant to Deleted: Tamper-Resistant Unlearning Through Weight-Space Regularization
  35. NeurIPS 2025ICLScan: Detecting Backdoors in Black-Box Large Language Models via Targeted In-context Illumination
  36. NeurIPS 2025Latency NMS Attacks: Is It Real Life or Is It Just Fantasy?
  37. NeurIPS 2025Lie Detector: Unified Backdoor Detection via Cross-Examination Framework
  38. NeurIPS 2025LoSplit: Loss-Guided Dynamic Split for Training-Time Defense Against Graph Backdoor Attacks
  39. NeurIPS 2025MARS: A Malignity-Aware Backdoor Defense in Federated Learning
  40. NeurIPS 2025Provable Watermarking for Data Poisoning Attacks
  41. NeurIPS 2025RepGuard: Adaptive Feature Decoupling for Robust Backdoor Defense in Large Language Models
  42. NeurIPS 2025SilentStriker: Toward Stealthy Bit-Flip Attacks on Large Language Models
  43. NeurIPS 2025SNEAKDOOR: Stealthy Backdoor Attacks against Distribution Matching-based Dataset Condensation
  44. NeurIPS 2025Stealthy Yet Effective: Distribution-Preserving Backdoor Attacks on Graph Classification
  45. NeurIPS 2025Taught Well Learned Ill: Towards Distillation-conditional Backdoor Attack
  46. NeurIPS 2025Temporal Logic-Based Multi-Vehicle Backdoor Attacks against Offline RL Agents in End-to-end Autonomous Driving
  47. NeurIPS 2025The Good, the Bad and the Ugly: Meta-Analysis of Watermarks, Transferable Attacks and Adversarial Defenses
  48. NeurIPS 2025ToxicTextCLIP: Text-Based Poisoning and Backdoor Attacks on CLIP Pre-training
  49. NeurIPS 2025Tracing Back the Malicious Clients in Poisoning Attacks to Federated Learning
  50. NeurIPS 2025TransferBench: Benchmarking Ensemble-based Black-box Transfer Attacks

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