对齐与安全论文趋势 2023–2026

主要结论

• 对齐与安全论文从 2023 年的 438 篇增长到 2026 年的 1,882 篇，三年间扩大到约 4.3 倍。
• NeurIPS 是发对齐与安全论文最多的会议，2023–2026 年共 1,421 篇，占四个会议总量的 31%。
• 交叉方向里涨得最快的是社会层面 · safety：同时属于两个主题的论文从 2023 年的 0 篇增至 2026 年的 137 篇。
• 在「对齐与负责任 AI」的 5 个主题里，它的 2026 年论文数排第 1；同组增长最快的社会层面 · safety是 93 倍，对齐与安全是 4.3 倍。

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

• NeurIPS 2026$\gamma$-weakly $\theta$-up-concavity: A Unified Framework for Non-Convex Optimization Beyond DR-Submodular and OSS Functions
• ICML 2026AICrypto: Evaluating Cryptography Capabilities of Large Language Models
• ICLR 2026A Fair Bayesian Inference through Matched Gibbs Posterior
• CVPR 2026Advancing Cancer Prognosis with Hierarchical Fusion of Genomic Proteomic and
• NeurIPS 2025$\Psi$-Sampler: Initial Particle Sampling for SMC-Based Inference-Time Reward Alignment in Score Models
• ICML 2025A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment
• ICLR 2025$\sigma$-zero: Gradient-based Optimization of $\ell_0$-norm Adversarial Examples
• CVPR 2025AesthetiQ: Enhancing Graphic Layout Design via Aesthetic-Aware Preference Alignment of Multi-modal Large Language Models
• NeurIPS 2024A survey and benchmark of high-dimensional Bayesian optimization of discrete sequences
• ICML 2024A Global Geometric Analysis of Maximal Coding Rate Reduction
• ICLR 2024A unique M-pattern for micro-expression spotting in long videos
• CVPR 2024Decompose-and-Compose: A Compositional Approach to Mitigating Spurious Correlation
• NeurIPS 2023(S)GD over Diagonal Linear Networks: Implicit bias, Large Stepsizes and Edge of Stability
• ICML 2023A Robust Optimisation Perspective on Counterexample-Guided Repair of Neural Networks
• ICLR 2023Aligning Model and Macaque Inferior Temporal Cortex Representations Improves Model-to-Human Behavioral Alignment and Adversarial Robustness
• CVPR 2023Adjustment and Alignment for Unbiased Open Set Domain Adaptation
• NeurIPS 2026$\mathcal{P}$Torch: Narrowing the Gap Between Projection and Gradient-Based Learning
• ICML 2026Alignment Pretraining: AI Discourse Causes Self-Fulfilling (Mis)alignment
• ICLR 2026A2D: Any-Order, Any-Step Safety Alignment for Diffusion Language Models
• CVPR 2026AlignPose Generalizable 6D Pose Estimation via Multi-view Feature-metric Alignment

全部对齐与安全论文

共 138 篇，第 2 / 3 页

51. NeurIPS 2023Is RLHF More Difficult than Standard RL? A Theoretical Perspective
52. NeurIPS 2023Jailbroken: How Does LLM Safety Training Fail?
53. NeurIPS 2023Language Model Alignment with Elastic Reset
54. NeurIPS 2023Language Quantized AutoEncoders: Towards Unsupervised Text-Image Alignment
55. NeurIPS 2023Large Language Models Are Zero-Shot Time Series Forecasters
56. NeurIPS 2023Latent Space Translation via Semantic Alignment
57. NeurIPS 2023Learning better with Dale’s Law: A Spectral Perspective
58. NeurIPS 2023Learning Fine-grained View-Invariant Representations from Unpaired Ego-Exo Videos via Temporal Alignment
59. NeurIPS 2023Learning from Active Human Involvement through Proxy Value Propagation
60. NeurIPS 2023Mass-Producing Failures of Multimodal Systems with Language Models
61. NeurIPS 2023Mind the spikes: Benign overfitting of kernels and neural networks in fixed dimension
62. NeurIPS 2023Minimax Forward and Backward Learning of Evolving Tasks with Performance Guarantees
63. NeurIPS 2023Minimum norm interpolation by perceptra: Explicit regularization and implicit bias
64. NeurIPS 2023MonoUNI: A Unified Vehicle and Infrastructure-side Monocular 3D Object Detection Network with Sufficient Depth Clues
65. NeurIPS 2023Necessary and Sufficient Conditions for Optimal Decision Trees using Dynamic Programming
66. NeurIPS 2023Neural (Tangent Kernel) Collapse
67. NeurIPS 2023Neural Harmonics: Bridging Spectral Embedding and Matrix Completion in Self-Supervised Learning
68. NeurIPS 2023On Learning Necessary and Sufficient Causal Graphs
69. NeurIPS 2023On Masked Pre-training and the Marginal Likelihood
70. NeurIPS 2023On Private and Robust Bandits
71. NeurIPS 2023On skip connections and normalisation layers in deep optimisation
72. NeurIPS 2023On the impact of activation and normalization in obtaining isometric embeddings at initialization
73. NeurIPS 2023On the Interplay between Social Welfare and Tractability of Equilibria
74. NeurIPS 2023On the Overlooked Pitfalls of Weight Decay and How to Mitigate Them: A Gradient-Norm Perspective
75. NeurIPS 2023OpenAssistant Conversations - Democratizing Large Language Model Alignment
76. NeurIPS 2023OpenProteinSet: Training data for structural biology at scale
77. NeurIPS 2023Optimal Guarantees for Algorithmic Reproducibility and Gradient Complexity in Convex Optimization
78. NeurIPS 2023Parallel Spiking Neurons with High Efficiency and Ability to Learn Long-term Dependencies
79. NeurIPS 2023Principle-Driven Self-Alignment of Language Models from Scratch with Minimal Human Supervision
80. NeurIPS 2023Private (Stochastic) Non-Convex Optimization Revisited: Second-Order Stationary Points and Excess Risks
81. NeurIPS 2023Quantus: An Explainable AI Toolkit for Responsible Evaluation of Neural Network Explanations and Beyond
82. NeurIPS 2023Rank-1 Matrix Completion with Gradient Descent and Small Random Initialization
83. NeurIPS 2023Red Teaming Deep Neural Networks with Feature Synthesis Tools
84. NeurIPS 2023Regret Matching+: (In)Stability and Fast Convergence in Games
85. NeurIPS 2023Residual Alignment: Uncovering the Mechanisms of Residual Networks
86. NeurIPS 2023Resolving the Tug-of-War: A Separation of Communication and Learning in Federated Learning
87. NeurIPS 2023Responsible AI (RAI) Games and Ensembles
88. NeurIPS 2023Revisiting Evaluation Metrics for Semantic Segmentation: Optimization and Evaluation of Fine-grained Intersection over Union
89. NeurIPS 2023Reward Finetuning for Faster and More Accurate Unsupervised Object Discovery
90. NeurIPS 2023Rewarded soups: towards Pareto-optimal alignment by interpolating weights fine-tuned on diverse rewards
91. NeurIPS 2023Robust covariance estimation with missing values and cell-wise contamination
92. NeurIPS 2023Robust Distributed Learning: Tight Error Bounds and Breakdown Point under Data Heterogeneity
93. NeurIPS 2023RRHF: Rank Responses to Align Language Models with Human Feedback
94. NeurIPS 2023Safe Exploration in Reinforcement Learning: A Generalized Formulation and Algorithms
95. NeurIPS 2023SafeDICE: Offline Safe Imitation Learning with Non-Preferred Demonstrations
96. NeurIPS 2023Safety Verification of Decision-Tree Policies in Continuous Time
97. NeurIPS 2023Sample Complexity for Quadratic Bandits: Hessian Dependent Bounds and Optimal Algorithms
98. NeurIPS 2023Self-supervised video pretraining yields robust and more human-aligned visual representations
99. NeurIPS 2023Sequential Predictive Two-Sample and Independence Testing
100. NeurIPS 2023Sequential Preference Ranking for Efficient Reinforcement Learning from Human Feedback

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。