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 300 篇，第 4 / 6 页

151. ICML 2025MindAligner: Explicit Brain Functional Alignment for Cross-Subject Visual Decoding from Limited fMRI Data
152. ICML 2025Mitigating Local Cohesion and Global Sparseness in Graph Contrastive Learning with Fuzzy Boundaries
153. ICML 2025Mixed-curvature decision trees and random forests
154. ICML 2025MM-RLHF: The Next Step Forward in Multimodal LLM Alignment
155. ICML 2025MMedPO: Aligning Medical Vision-Language Models with Clinical-Aware Multimodal Preference Optimization
156. ICML 2025MONA: Myopic Optimization with Non-myopic Approval Can Mitigate Multi-step Reward Hacking
157. ICML 2025MPO: An Efficient Post-Processing Framework for Mixing Diverse Preference Alignment
158. ICML 2025Natural Perturbations for Black-box Training of Neural Networks by Zeroth-Order Optimization
159. ICML 2025Nesterov Method for Asynchronous Pipeline Parallel Optimization
160. ICML 2025Neural Collapse Beyond the Unconstrained Features Model: Landscape, Dynamics, and Generalization in the Mean-Field Regime
161. ICML 2025Neural Representational Consistency Emerges from Probabilistic Neural-Behavioral Representation Alignment
162. ICML 2025Neural Solver Selection for Combinatorial Optimization
163. ICML 2025No Task Left Behind: Isotropic Model Merging with Common and Task-Specific Subspaces
164. ICML 2025Non-Asymptotic and Non-Lipschitzian Bounds on Optimal Values in Stochastic Optimization Under Heavy Tails
165. ICML 2025Novelty Detection in Reinforcement Learning with World Models
166. ICML 2025Observation Interference in Partially Observable Assistance Games
167. ICML 2025Of Mice and Machines: A Comparison of Learning Between Real World Mice and RL Agents
168. ICML 2025Offline Model-based Optimization for Real-World Molecular Discovery
169. ICML 2025On Expressive Power of Looped Transformers: Theoretical Analysis and Enhancement via Timestep Encoding
170. ICML 2025On Linear Convergence in Smooth Convex-Concave Bilinearly-Coupled Saddle-Point Optimization: Lower Bounds and Optimal Algorithms
171. ICML 2025On Teacher Hacking in Language Model Distillation
172. ICML 2025On the Alignment between Fairness and Accuracy: from the Perspective of Adversarial Robustness
173. ICML 2025On the Robustness of Reward Models for Language Model Alignment
174. ICML 2025On the Similarities of Embeddings in Contrastive Learning
175. ICML 2025On the Tension between Byzantine Robustness and No-Attack Accuracy in Distributed Learning
176. ICML 2025On the Vulnerability of Applying Retrieval-Augmented Generation within Knowledge-Intensive Application Domains
177. ICML 2025One-Step Generalization Ratio Guided Optimization for Domain Generalization
178. ICML 2025Optimal Sensor Scheduling and Selection for Continuous-Discrete Kalman Filtering with Auxiliary Dynamics
179. ICML 2025Optimization Proxies using Limited Labeled Data and Training Time -- A Semi-Supervised Bayesian Neural Network Approach
180. ICML 2025OR-Bench: An Over-Refusal Benchmark for Large Language Models
181. ICML 2025PANDAS: Improving Many-shot Jailbreaking via Positive Affirmation, Negative Demonstration, and Adaptive Sampling
182. ICML 2025Pareto Merging: Multi-Objective Optimization for Preference-Aware Model Merging
183. ICML 2025PARM: Multi-Objective Test-Time Alignment via Preference-Aware Autoregressive Reward Model
184. ICML 2025PILAF: Optimal Human Preference Sampling for Reward Modeling
185. ICML 2025PIPA: Preference Alignment as Prior-Informed Statistical Estimation
186. ICML 2025PipeOffload: Improving Scalability of Pipeline Parallelism with Memory Optimization
187. ICML 2025PoisonBench: Assessing Language Model Vulnerability to Poisoned Preference Data
188. ICML 2025Policy Filtration for RLHF to Mitigate Noise in Reward Models
189. ICML 2025Policy Optimization for CMDPs with Bandit Feedback: Learning Stochastic and Adversarial Constraints
190. ICML 2025Policy-labeled Preference Learning: Is Preference Enough for RLHF?
191. ICML 2025POROver: Improving Safety and Reducing Overrefusal in Large Language Models with Overgeneration and Preference Optimization
192. ICML 2025Position: AI Safety Must Embrace an Antifragile Perspective
193. ICML 2025Position: AI Safety should prioritize the Future of Work
194. ICML 2025Position: Beyond Assistance – Reimagining LLMs as Ethical and Adaptive Co-Creators in Mental Health Care
195. ICML 2025Position: Challenges and Future Directions of Data-Centric AI Alignment
196. ICML 2025Position: Current Model Licensing Practices are Dragging Us into a Quagmire of Legal Noncompliance
197. ICML 2025Position: Editing Large Language Models Poses Serious Safety Risks
198. ICML 2025Position: Humanity Faces Existential Risk from Gradual Disempowerment
199. ICML 2025Position: Iterative Online-Offline Joint Optimization is Needed to Manage Complex LLM Copyright Risks
200. ICML 2025Position: Political Neutrality in AI Is Impossible — But Here Is How to Approximate It

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。