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 61 / 92 页

  1. NeurIPS 2025Intermediate Domain Alignment and Morphology Analogy for Patent-Product Image Retrieval
  2. NeurIPS 2025InterMT: Multi-Turn Interleaved Preference Alignment with Human Feedback
  3. NeurIPS 2025IPAD: Inverse Prompt for AI Detection - A Robust and Interpretable LLM-Generated Text Detector
  4. NeurIPS 2025JADE: Joint Alignment and Deep Embedding for Multi-Slice Spatial Transcriptomics
  5. NeurIPS 2025JailBound: Jailbreaking Internal Safety Boundaries of Vision-Language Models
  6. NeurIPS 2025Keeping an Eye on LLM Unlearning: The Hidden Risk and Remedy
  7. NeurIPS 2025KL Penalty Control via Perturbation for Direct Preference Optimization
  8. NeurIPS 2025KL-Regularized RLHF with Multiple Reference Models: Exact Solutions and Sample Complexity
  9. NeurIPS 2025LABridge: Text–Image Latent Alignment Framework via Mean-Conditioned OU Process
  10. NeurIPS 2025Language Model Behavioral Phases are Consistent Across Architecture, Training Data, and Scale
  11. NeurIPS 2025Language Models Are Capable of Metacognitive Monitoring and Control of Their Internal Activations
  12. NeurIPS 2025Large language models can learn and generalize steganographic chain-of-thought under process supervision
  13. NeurIPS 2025LARGO: Latent Adversarial Reflection through Gradient Optimization for Jailbreaking LLMs
  14. NeurIPS 2025LASeR: Learning to Adaptively Select Reward Models with Multi-Arm Bandits
  15. NeurIPS 2025Latent Harmony: Synergistic Unified UHD Image Restoration via Latent Space Regularization and Controllable Refinement
  16. NeurIPS 2025LayerNavigator: Finding Promising Intervention Layers for Efficient Activation Steering in Large Language Models
  17. NeurIPS 2025Learning “Partner-Aware” Collaborators in Multi-Party Collaboration
  18. NeurIPS 2025Learning a Cross-Modal Schrödinger Bridge for Visual Domain Generalization
  19. NeurIPS 2025Learning single index models via harmonic decomposition
  20. NeurIPS 2025Learning to Add, Multiply, and Execute Algorithmic Instructions Exactly with Neural Networks
  21. NeurIPS 2025Learning to Steer: Input-dependent Steering for Multimodal LLMs
  22. NeurIPS 2025Less is More: Improving LLM Alignment via Preference Data Selection
  23. NeurIPS 2025Let a Neural Network be Your Invariant
  24. NeurIPS 2025Leveraging Importance Sampling to Detach Alignment Modules from Large Language Models
  25. NeurIPS 2025Leveraging robust optimization for llm alignment under distribution shifts
  26. NeurIPS 2025LeVo: High-Quality Song Generation with Multi-Preference Alignment
  27. NeurIPS 2025Lifelong Safety Alignment for Language Models
  28. NeurIPS 2025LLM Safety Alignment is Divergence Estimation in Disguise
  29. NeurIPS 2025LLMs Encode Harmfulness and Refusal Separately
  30. NeurIPS 2025LongVPO: From Anchored Cues to Self-Reasoning for Long-Form Video Preference Optimization
  31. NeurIPS 2025Machine Unlearning Doesn't Do What You Think: Lessons for Generative AI Policy and Research
  32. NeurIPS 2025macOSWorld: A Multilingual Interactive Benchmark for GUI Agents
  33. NeurIPS 2025Many LLMs Are More Utilitarian Than One
  34. NeurIPS 2025Matchings Under Biased and Correlated Evaluations
  35. NeurIPS 2025Measuring and Guiding Monosemanticity
  36. NeurIPS 2025Mechanism Design for LLM Fine-tuning with Multiple Reward Models
  37. NeurIPS 2025Median Selection with Noisy and Structural Information
  38. NeurIPS 2025Mesh-RFT: Enhancing Mesh Generation via Fine-grained Reinforcement Fine-Tuning
  39. NeurIPS 2025Meta-Learning Objectives for Preference Optimization
  40. NeurIPS 2025MGUP: A Momentum-Gradient Alignment Update Policy for Stochastic Optimization
  41. NeurIPS 2025Military AI Needs Technically-Informed Regulation to Safeguard AI Research and its Applications
  42. NeurIPS 2025Mitigating Hallucination Through Theory-Consistent Symmetric Multimodal Preference Optimization
  43. NeurIPS 2025Mitigating Reward Over-optimization in Direct Alignment Algorithms with Importance Sampling
  44. NeurIPS 2025Mitigating Sexual Content Generation via Embedding Distortion in Text-conditioned Diffusion Models
  45. NeurIPS 2025Mix Data or Merge Models? Balancing the Helpfulness, Honesty, and Harmlessness of Large Language Model via Model Merging
  46. NeurIPS 2025MJ-Video: Benchmarking and Rewarding Video Generation with Fine-Grained Video Preference
  47. NeurIPS 2025Model Inversion with Layer-Specific Modeling and Alignment for Data-Free Continual Learning
  48. NeurIPS 2025Model–Behavior Alignment under Flexible Evaluation: When the Best-Fitting Model Isn’t the Right One
  49. NeurIPS 2025Multi-step Visual Reasoning with Visual Tokens Scaling and Verification
  50. NeurIPS 2025Multimodal Disease Progression Modeling via Spatiotemporal Disentanglement and Multiscale 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