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 480 篇，第 6 / 10 页

251. NeurIPS 2026MIND-DDI: Multi-Omics Interpretable Drug-Drug Interaction Prediction with Joint Optimization of Graph Structure, Neural Architecture, and Symbolic Rules
252. NeurIPS 2026MindGuard: Guardrail Classifiers for Multi-Turn Mental Health Support
253. NeurIPS 2026MLLMEraser: Achieving Test-Time Unlearning in Multimodal Large Language Models through Activation Steering
254. NeurIPS 2026MMA-SafetyBench: A Benchmark for Multimodal Agent Safety Evaluation
255. NeurIPS 2026Model Capacity Determines Grokking through Competing Memorisation and Generalisation Speeds
256. NeurIPS 2026Model Immunization Beyond Condition Numbers: The Importance of Plateau Regions
257. NeurIPS 2026Models That Know How Evaluations Are Designed Score Safer
258. NeurIPS 2026MorphGen: Controllable Cell-Image Generation with Biological Representation Alignment
259. NeurIPS 2026Mult-DPO: Multinomial Direct Preference Optimization for Recommender Systems
260. NeurIPS 2026Multi-Level Alignment Framework for Long-Term Olfactory Neural Decoding
261. NeurIPS 2026Multilingual Safety Alignment via Self-Distillation
262. NeurIPS 2026MUTE: Multi-Level Alignment Uncoupling Against Talking-Head Exploitation for Voice Protection
263. NeurIPS 2026Near-Optimal Last-Iterate Convergence for Zero-Sum Games with Bandit Feedback and Opponent Actions
264. NeurIPS 2026Negation Neglect: When models fail to learn negations in training
265. NeurIPS 2026Neural Chameleons: Language Models Can Learn to Hide Their Thoughts from Unseen Activation Monitors
266. NeurIPS 2026Noisy isomorphism: Robust expressivity of GNNs under structural perturbation
267. NeurIPS 2026Noisy2Latent: Nonparametric Deconvolution and Denoising via Maximum Mean Discrepancy
268. NeurIPS 2026Nonlinear Direct Feedback Alignment for Scalable Backpropagation-Free Training
269. NeurIPS 2026Not all uncertainty is alike: volatility, stochasticity, and exploration
270. NeurIPS 2026Not Too Generative, Not Too Discriminative: The Human Alignment Sweet Spot
271. NeurIPS 2026OBLIQ-Bench: Exposing Overlooked Bottlenecks in Modern Retrievers with Latent and Implicit Queries
272. NeurIPS 2026Observations Drift, Structures Remain: Structural Pretraining with Time Alignment for Electromagnetic Signals
273. NeurIPS 2026Omni-Safety under Cross-Modality Conflict: Vulnerabilities, Dynamic Mechanisms and Efficient Alignment
274. NeurIPS 2026On approximation and estimation of Schrödinger potentials without the curse of dimensionality
275. NeurIPS 2026On Lipschitz Explosion in Deep Neural Networks with Normalization: Consequences for Optimization and Robustness
276. NeurIPS 2026On the Depth of Monotone ReLU Neural Networks and ICNNs
277. NeurIPS 2026On the Parallel Optimality of Exponentiated Gradient Descent
278. NeurIPS 2026On the Sparsity of Direct Preference Optimization: Weight Disentanglement in the NTK Regime
279. NeurIPS 2026On the Tightness and Computational Tractability of Higher-Dimensional Confidence Sequences
280. NeurIPS 2026On-Policy Consistency Training Improves LLM Safety with Minimal Capability Degradation
281. NeurIPS 2026One-Time Soft Alignment Enables Resilient Learning without Weight Transport
282. NeurIPS 2026OperatorSHAP: Fast and Accurate Shapley Value Estimation for Neural Operators
283. NeurIPS 2026Optimal and Efficient Contextual Combinatorial Semi-bandits with General Function Approximation
284. NeurIPS 2026Optimal In-context Adaptivity and Distributional Robustness of Transformers
285. NeurIPS 2026Optimal Representation Size: High-Dimensional Analysis of Pretraining and Linear Probing
286. NeurIPS 2026Optimal Transport Reweighting for Robust Learning under Spurious Correlations and Label Noise
287. NeurIPS 2026Oracle-Robust Online Alignment for Large Language Models
288. NeurIPS 2026ORCA: Orthogonal Residual Consensus Alignment for Multi-View Clustering
289. NeurIPS 2026Orthogonal Sparse Subgraph Alignment for Structure-Function Coupling in Brain Networks
290. NeurIPS 2026Overcoming Rank Collapse in Feedback Alignment
291. NeurIPS 2026Pairwise AUC Optimization Needs Corrective Power: A Unified View
292. NeurIPS 2026Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs
293. NeurIPS 2026PaperLens: How Predictable Is Paper Acceptance?
294. NeurIPS 2026ParetoM$^3$: Learning on the Pareto Set under Preference Guidance via Min-Max-Min Optimization
295. NeurIPS 2026Participatory ML for Social Harm Should be Constructed, Validated and Reasoned Through First-Person Accounts
296. NeurIPS 2026Persona Vectors: Monitoring and Controlling Character Traits in Language Models
297. NeurIPS 2026Personalized Safety in Federated Fine-Tuning of Large Language Models
298. NeurIPS 2026PISG: Constraint-Aligned Signal Amplification for Diffusion-Based Combinatorial Optimization
299. NeurIPS 2026PixelDense: Dense Prediction as Representation Alignment for Pixel Diffusion
300. NeurIPS 2026POME: Post Optimization Model Edit via Muon-Style Projection

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。