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 1,339 篇，第 3 / 27 页

101. ICLR 2026Closing the Modality Gap Aligns Group-Wise Semantics
102. ICLR 2026Closing the Safety Gap: Surgical Concept Erasure in Visual Autoregressive Models
103. ICLR 2026Co-occurring Associated REtained concepts in Diffusion Unlearning
104. ICLR 2026CoDA: From Text-to-Image Diffusion Models to Training-Free Dataset Distillation
105. ICLR 2026CodeGenGuard: A Robust Watermark for Code Generation Models
106. ICLR 2026CoFact: Conformal Factuality Guarantees for Language Models under Distribution Shift
107. ICLR 2026CogFlow: Bridging Perception and Reasoning through Knowledge Internalization for Visual Mathematical Problem Solving
108. ICLR 2026COMAL: A Convergent Meta-Algorithm for Aligning LLMs with General Preferences
109. ICLR 2026Completed Hyperparameter Transfer across Modules, Width, Depth, Batch and Duration
110. ICLR 2026Concept-Aware Privacy Mechanisms for Defending Embedding Inversion Attacks
111. ICLR 2026ConRep4CO: Contrastive Representation Learning of Combinatorial Optimization Instances across Types
112. ICLR 2026Consis-GCPO: Consistency-Preserving Group Causal Preference Optimization for Vision Customization
113. ICLR 2026Consistent Noisy Latent Rewards for Trajectory Preference Optimization in Diffusion Models
114. ICLR 2026Constitutional Classifiers++: Production-Grade Defenses against Universal Jailbreaks
115. ICLR 2026ContextBench: Modifying Contexts for Targeted Latent Activation and Behaviour Elicitation
116. ICLR 2026Control Tax: The Price of Keeping AI in Check
117. ICLR 2026Convergence Analysis of Tsetlin Machines under Noise-Free and Noisy Training Conditions: from $2$ Bits to $k$ Bits
118. ICLR 2026Convergent Differential Privacy Analysis for General Federated Learning
119. ICLR 2026Convex Dominance in Deep Learning I: A Scaling Law of Loss and Learning Rate
120. ICLR 2026Counterfactual LLM-based Framework for Measuring Rhetorical Style
121. ICLR 2026Critical Confabulation: Can LLMs Hallucinate for Social Good?
122. ICLR 2026Cultivating Pluralism In Algorithmic Monoculture: The Community Alignment Dataset
123. ICLR 2026Curation Leaks: Membership Inference Attacks against Data Curation for Machine Learning
124. ICLR 2026D&R: Recovery-based AI-Generated Text Detection via a Single Black-box LLM Call
125. ICLR 2026Data Aware and Scalable Sensitivity Analysis for Decision Tree Ensembles
126. ICLR 2026Data Selection for LLM Alignment Using Fine-Grained Preferences
127. ICLR 2026DecAlign: Hierarchical Cross-Modal Alignment for Decoupled Multimodal Representation Learning
128. ICLR 2026Decomposing Extrapolative Problem Solving: Spatial Transfer and Length Scaling with Map Worlds
129. ICLR 2026Decoupling Primitive with Experts: Dynamic Feature Alignment for Compositional Zero-Shot Learning
130. ICLR 2026Deep Ignorance: Filtering Pretraining Data Builds Tamper-Resistant Safeguards into Open-Weight LLMs
131. ICLR 2026Deep Latent Variable Model based Vertical Federated Learning with Flexible Alignment and Labeling Scenarios
132. ICLR 2026Defending against Backdoor Attacks via Module Switching
133. ICLR 2026DeRaDiff: Denoising Time Realignment of Diffusion Models
134. ICLR 2026Developmental Federated Tuning: A Cognitive-Inspired Paradigm for Efficient LLM Adaptation
135. ICLR 2026Differentially Private Domain Discovery
136. ICLR 2026DiffuGuard: How Intrinsic Safety is Lost and Found in Diffusion Large Language Models
137. ICLR 2026Diffusion Alignment as Variataional Expectation-Maximization
138. ICLR 2026Diffusion Negative Preference Optimization Made Simple
139. ICLR 2026Direct Preference Optimization for Primitive-Enabled Hierarchical RL: A Bilevel Approach
140. ICLR 2026Direct Reward Fine-Tuning on Poses for Single Image to 3D Human in the Wild
141. ICLR 2026Discern Truth from Falsehood: Reducing Over-Refusal via Contrastive Refinement
142. ICLR 2026Discrete Diffusion Trajectory Alignment via Stepwise Decomposition
143. ICLR 2026Disentangling Length Bias in Preference Learning via Response-Conditioned Modeling
144. ICLR 2026Disentangling the Factors of Convergence between Brains and Computer Vision Models
145. ICLR 2026Displacement-Resistant Extensions of DPO with Nonconvex $f$-Divergences
146. ICLR 2026DistDF: Time-series Forecasting Needs Joint-distribution Wasserstein Alignment
147. ICLR 2026Distilling the Thought, Watermarking the Answer: A Principle Semantic Guided Watermark for Reasoning Large Language Models
148. ICLR 2026Distributional Vision-Language Alignment by Cauchy-Schwarz Divergence
149. ICLR 2026DIVERSE: Disagreement-Inducing Vector Evolution for Rashomon Set Exploration
150. ICLR 2026Do Large Language Models Know What They Are Capable Of?

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。