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 90 / 92 页

  1. NeurIPS 2023Efficient Data Subset Selection to Generalize Training Across Models: Transductive and Inductive Networks
  2. NeurIPS 2023Exact Verification of ReLU Neural Control Barrier Functions
  3. NeurIPS 2023Expert load matters: operating networks at high accuracy and low manual effort
  4. NeurIPS 2023Explainable and Efficient Randomized Voting Rules
  5. NeurIPS 2023Extraction and Recovery of Spatio-Temporal Structure in Latent Dynamics Alignment with Diffusion Models
  6. NeurIPS 2023Fast and Simple Spectral Clustering in Theory and Practice
  7. NeurIPS 2023FD-Align: Feature Discrimination Alignment for Fine-tuning Pre-Trained Models in Few-Shot Learning
  8. NeurIPS 2023Feature learning via mean-field Langevin dynamics: classifying sparse parities and beyond
  9. NeurIPS 2023Framework and Benchmarks for Combinatorial and Mixed-variable Bayesian Optimization
  10. NeurIPS 2023GALOPA: Graph Transport Learning with Optimal Plan Alignment
  11. NeurIPS 2023Generalized equivalences between subsampling and ridge regularization
  12. NeurIPS 2023GeoCLIP: Clip-Inspired Alignment between Locations and Images for Effective Worldwide Geo-localization
  13. NeurIPS 2023Graph Neural Networks for Road Safety Modeling: Datasets and Evaluations for Accident Analysis
  14. NeurIPS 2023Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery
  15. NeurIPS 2023Honesty Is the Best Policy: Defining and Mitigating AI Deception
  16. NeurIPS 2023Imitation Learning from Imperfection: Theoretical Justifications and Algorithms
  17. NeurIPS 2023Inference-Time Intervention: Eliciting Truthful Answers from a Language Model
  18. NeurIPS 2023Initialization-Dependent Sample Complexity of Linear Predictors and Neural Networks
  19. NeurIPS 2023Interpretability at Scale: Identifying Causal Mechanisms in Alpaca
  20. NeurIPS 2023IPMix: Label-Preserving Data Augmentation Method for Training Robust Classifiers
  21. NeurIPS 2023Is RLHF More Difficult than Standard RL? A Theoretical Perspective
  22. NeurIPS 2023Jailbroken: How Does LLM Safety Training Fail?
  23. NeurIPS 2023Language Model Alignment with Elastic Reset
  24. NeurIPS 2023Language Quantized AutoEncoders: Towards Unsupervised Text-Image Alignment
  25. NeurIPS 2023Large Language Models Are Zero-Shot Time Series Forecasters
  26. NeurIPS 2023Latent Space Translation via Semantic Alignment
  27. NeurIPS 2023Learning better with Dale’s Law: A Spectral Perspective
  28. NeurIPS 2023Learning Fine-grained View-Invariant Representations from Unpaired Ego-Exo Videos via Temporal Alignment
  29. NeurIPS 2023Learning from Active Human Involvement through Proxy Value Propagation
  30. NeurIPS 2023Mass-Producing Failures of Multimodal Systems with Language Models
  31. NeurIPS 2023Mind the spikes: Benign overfitting of kernels and neural networks in fixed dimension
  32. NeurIPS 2023Minimax Forward and Backward Learning of Evolving Tasks with Performance Guarantees
  33. NeurIPS 2023Minimum norm interpolation by perceptra: Explicit regularization and implicit bias
  34. NeurIPS 2023MonoUNI: A Unified Vehicle and Infrastructure-side Monocular 3D Object Detection Network with Sufficient Depth Clues
  35. NeurIPS 2023Necessary and Sufficient Conditions for Optimal Decision Trees using Dynamic Programming
  36. NeurIPS 2023Neural (Tangent Kernel) Collapse
  37. NeurIPS 2023Neural Harmonics: Bridging Spectral Embedding and Matrix Completion in Self-Supervised Learning
  38. NeurIPS 2023On Learning Necessary and Sufficient Causal Graphs
  39. NeurIPS 2023On Masked Pre-training and the Marginal Likelihood
  40. NeurIPS 2023On Private and Robust Bandits
  41. NeurIPS 2023On skip connections and normalisation layers in deep optimisation
  42. NeurIPS 2023On the impact of activation and normalization in obtaining isometric embeddings at initialization
  43. NeurIPS 2023On the Interplay between Social Welfare and Tractability of Equilibria
  44. NeurIPS 2023On the Overlooked Pitfalls of Weight Decay and How to Mitigate Them: A Gradient-Norm Perspective
  45. NeurIPS 2023OpenAssistant Conversations - Democratizing Large Language Model Alignment
  46. NeurIPS 2023OpenProteinSet: Training data for structural biology at scale
  47. NeurIPS 2023Optimal Guarantees for Algorithmic Reproducibility and Gradient Complexity in Convex Optimization
  48. NeurIPS 2023Parallel Spiking Neurons with High Efficiency and Ability to Learn Long-term Dependencies
  49. NeurIPS 2023Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervision
  50. NeurIPS 2023Private (Stochastic) Non-Convex Optimization Revisited: Second-Order Stationary Points and Excess Risks

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