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 875 篇，第 5 / 18 页

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

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