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 70 / 92 页

  1. ICLR 2024Making RL with Preference-based Feedback Efficient via Randomization
  2. ICLR 2024MAP IT to Visualize Representations
  3. ICLR 2024Masks, Signs, And Learning Rate Rewinding
  4. ICLR 2024MINT: Evaluating LLMs in Multi-turn Interaction with Tools and Language Feedback
  5. ICLR 2024Motif: Intrinsic Motivation from Artificial Intelligence Feedback
  6. ICLR 2024Multilingual Jailbreak Challenges in Large Language Models
  7. ICLR 2024NEFTune: Noisy Embeddings Improve Instruction Finetuning
  8. ICLR 2024Network Memory Footprint Compression Through Jointly Learnable Codebooks and Mappings
  9. ICLR 2024Neuron Activation Coverage: Rethinking Out-of-distribution Detection and Generalization
  10. ICLR 2024Neuron-Enhanced AutoEncoder Matrix Completion and Collaborative Filtering: Theory and Practice
  11. ICLR 2024On Bias-Variance Alignment in Deep Models
  12. ICLR 2024On the Fairness ROAD: Robust Optimization for Adversarial Debiasing
  13. ICLR 2024On the Generalization and Approximation Capacities of Neural Controlled Differential Equations
  14. ICLR 2024On the Over-Memorization During Natural, Robust and Catastrophic Overfitting
  15. ICLR 2024On-Policy Distillation of Language Models: Learning from Self-Generated Mistakes
  16. ICLR 2024Optimal transport based adversarial patch to leverage large scale attack transferability
  17. ICLR 2024Overthinking the Truth: Understanding how Language Models Process False Demonstrations
  18. ICLR 2024PARL: A Unified Framework for Policy Alignment in Reinforcement Learning from Human Feedback
  19. ICLR 2024Plug-and-Play: An Efficient Post-training Pruning Method for Large Language Models
  20. ICLR 2024Privately Aligning Language Models with Reinforcement Learning
  21. ICLR 2024Quantifying and Enhancing Multi-modal Robustness with Modality Preference
  22. ICLR 2024Query-Dependent Prompt Evaluation and Optimization with Offline Inverse RL
  23. ICLR 2024Quick-Tune: Quickly Learning Which Pretrained Model to Finetune and How
  24. ICLR 2024RAIN: Your Language Models Can Align Themselves without Finetuning
  25. ICLR 2024RECOMBINER: Robust and Enhanced Compression with Bayesian Implicit Neural Representations
  26. ICLR 2024Respect the model: Fine-grained and Robust Explanation with Sharing Ratio Decomposition
  27. ICLR 2024Revisit and Outstrip Entity Alignment: A Perspective of Generative Models
  28. ICLR 2024Reward Model Ensembles Help Mitigate Overoptimization
  29. ICLR 2024Ring-A-Bell! How Reliable are Concept Removal Methods For Diffusion Models?
  30. ICLR 2024RLCD: Reinforcement Learning from Contrastive Distillation for LM Alignment
  31. ICLR 2024Robust Similarity Learning with Difference Alignment Regularization
  32. ICLR 2024Robustifying and Boosting Training-Free Neural Architecture Search
  33. ICLR 2024Safe and Robust Watermark Injection with a Single OoD Image
  34. ICLR 2024Safe RLHF: Safe Reinforcement Learning from Human Feedback
  35. ICLR 2024Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructions
  36. ICLR 2024SALMON: Self-Alignment with Instructable Reward Models
  37. ICLR 2024Scaling for Training Time and Post-hoc Out-of-distribution Detection Enhancement
  38. ICLR 2024SCHEMA: State CHangEs MAtter for Procedure Planning in Instructional Videos
  39. ICLR 2024Self-Alignment with Instruction Backtranslation
  40. ICLR 2024Self-contradictory Hallucinations of Large Language Models: Evaluation, Detection and Mitigation
  41. ICLR 2024Self-supervised Pocket Pretraining via Protein Fragment-Surroundings Alignment
  42. ICLR 2024Set Learning for Accurate and Calibrated Models
  43. ICLR 2024SliceGPT: Compress Large Language Models by Deleting Rows and Columns
  44. ICLR 2024Sparse Spiking Neural Network: Exploiting Heterogeneity in Timescales for Pruning Recurrent SNN
  45. ICLR 2024Statistical Rejection Sampling Improves Preference Optimization
  46. ICLR 2024Subtractive Mixture Models via Squaring: Representation and Learning
  47. ICLR 2024Sudden Drops in the Loss: Syntax Acquisition, Phase Transitions, and Simplicity Bias in MLMs
  48. ICLR 2024Task structure and nonlinearity jointly determine learned representational geometry
  49. ICLR 2024Teaching Language Models to Hallucinate Less with Synthetic Tasks
  50. ICLR 2024The Alignment Problem from a Deep Learning Perspective

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