对齐与安全论文趋势 2023–2026

RLHF、偏好优化、红队与安全对齐

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 4,558 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

对齐与安全论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
202313887101112438
2024348156246125875
20254554733001351,363
20264806235512281,882

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

全部对齐与安全论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 4,558 篇，第 43 / 92 页

  1. ICLR 2025Circuit Representation Learning with Masked Gate Modeling and Verilog-AIG Alignment
  2. ICLR 2025Collab: Controlled Decoding using Mixture of Agents for LLM Alignment
  3. ICLR 2025CollabEdit: Towards Non-destructive Collaborative Knowledge Editing
  4. ICLR 2025Collapsed Language Models Promote Fairness
  5. ICLR 2025Comparing Targeting Strategies for Maximizing Social Welfare with Limited Resources
  6. ICLR 2025Complementary Label Learning with Positive Label Guessing and Negative Label Enhancement
  7. ICLR 2025Concept-ROT: Poisoning Concepts in Large Language Models with Model Editing
  8. ICLR 2025ConcreTizer: Model Inversion Attack via Occupancy Classification and Dispersion Control for 3D Point Cloud Restoration
  9. ICLR 2025Confidence Elicitation: A New Attack Vector for Large Language Models
  10. ICLR 2025Conformal Prediction Sets Can Cause Disparate Impact
  11. ICLR 2025Conformalized Interactive Imitation Learning: Handling Expert Shift and Intermittent Feedback
  12. ICLR 2025Conservative Contextual Bandits: Beyond Linear Representations
  13. ICLR 2025Context-Alignment: Activating and Enhancing LLMs Capabilities in Time Series
  14. ICLR 2025Continuous Autoregressive Modeling with Stochastic Monotonic Alignment for Speech Synthesis
  15. ICLR 2025Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements
  16. ICLR 2025Controllable Satellite-to-Street-View Synthesis with Precise Pose Alignment and Zero-Shot Environmental Control
  17. ICLR 2025Controllable Unlearning for Image-to-Image Generative Models via $\epsilon$-Constrained Optimization
  18. ICLR 2025Convergence and Implicit Bias of Gradient Descent on Continual Linear Classification
  19. ICLR 2025Convergent Privacy Loss of Noisy-SGD without Convexity and Smoothness
  20. ICLR 2025Copyright-Protected Language Generation via Adaptive Model Fusion
  21. ICLR 2025Correcting the Mythos of KL-Regularization: Direct Alignment without Overoptimization via Chi-Squared Preference Optimization
  22. ICLR 2025Correlated Proxies: A New Definition and Improved Mitigation for Reward Hacking
  23. ICLR 2025CREAM: Consistency Regularized Self-Rewarding Language Models
  24. ICLR 2025Cross-Modal Safety Mechanism Transfer in Large Vision-Language Models
  25. ICLR 2025Ctrl-U: Robust Conditional Image Generation via Uncertainty-aware Reward Modeling
  26. ICLR 2025DailyDilemmas: Revealing Value Preferences of LLMs with Quandaries of Daily Life
  27. ICLR 2025DarkBench: Benchmarking Dark Patterns in Large Language Models
  28. ICLR 2025Data Distillation for extrapolative protein design through exact preference optimization
  29. ICLR 2025Data Taggants: Dataset Ownership Verification Via Harmless Targeted Data Poisoning
  30. ICLR 2025Data-adaptive Differentially Private Prompt Synthesis for In-Context Learning
  31. ICLR 2025Dataset Ownership Verification in Contrastive Pre-trained Models
  32. ICLR 2025DCT-CryptoNets: Scaling Private Inference in the Frequency Domain
  33. ICLR 2025Deep Kernel Relative Test for Machine-generated Text Detection
  34. ICLR 2025Democratic Training Against Universal Adversarial Perturbations
  35. ICLR 2025Detecting Backdoor Samples in Contrastive Language Image Pretraining
  36. ICLR 2025Differentiable Optimization of Similarity Scores Between Models and Brains
  37. ICLR 2025Differentially Private Federated Learning with Time-Adaptive Privacy Spending
  38. ICLR 2025Differentially private optimization for non-decomposable objective functions
  39. ICLR 2025Differentially Private Steering for Large Language Model Alignment
  40. ICLR 2025Diffusion-NPO: Negative Preference Optimization for Better Preference Aligned Generation of Diffusion Models
  41. ICLR 2025DiffusionGuard: A Robust Defense Against Malicious Diffusion-based Image Editing
  42. ICLR 2025Direct Post-Training Preference Alignment for Multi-Agent Motion Generation Model Using Implicit Feedback from Pre-training Demonstrations
  43. ICLR 2025DiSK: Differentially Private Optimizer with Simplified Kalman Filter for Noise Reduction
  44. ICLR 2025Dissecting Adversarial Robustness of Multimodal LM Agents
  45. ICLR 2025Diverse Preference Learning for Capabilities and Alignment
  46. ICLR 2025Do as I do (Safely): Mitigating Task-Specific Fine-tuning Risks in Large Language Models
  47. ICLR 2025DocMIA: Document-Level Membership Inference Attacks against DocVQA Models
  48. ICLR 2025Does Refusal Training in LLMs Generalize to the Past Tense?
  49. ICLR 2025Does Safety Training of LLMs Generalize to Semantically Related Natural Prompts?
  50. ICLR 2025DSPO: Direct Score Preference Optimization for Diffusion Model Alignment

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 查对齐与安全在 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": "alignment-and-safety"}

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对齐与安全论文？
共 4,558 篇，按年份分别是 2023 年 438 篇、2024 年 875 篇、2025 年 1,363 篇、2026 年 1,882 篇。
NeurIPS 2026 有多少篇对齐与安全论文？
NeurIPS 2026 有 480 篇对齐与安全论文，上一届是 455 篇。
ICLR 2026 有多少篇对齐与安全论文？
ICLR 2026 有 623 篇对齐与安全论文，上一届是 473 篇。
ICML 2026 有多少篇对齐与安全论文？
ICML 2026 有 551 篇对齐与安全论文，上一届是 300 篇。
CVPR 2026 有多少篇对齐与安全论文？
CVPR 2026 有 228 篇对齐与安全论文，上一届是 135 篇。
哪个会议的对齐与安全论文最多？
NeurIPS，2023–2026 年共 1,421 篇。

「对齐与负责任 AI」下的其他主题