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 1,421 篇，第 10 / 29 页

451. NeurIPS 2026Uncovering Hidden Propensities in Language Models via Limited-Parameter Finetuning
452. NeurIPS 2026Uncovering the latent structure of interwoven population and temporal codes
453. NeurIPS 2026Understanding and Mitigating Under-Confidence in GNNs from the Final Layer
454. NeurIPS 2026Uniform Spectral Growth under Factor-wise Muon Orthogonalization in Matrix Factorization and LoRA
455. NeurIPS 2026Uniform Stability and Generalization Error of GD and SGD on Fixed-Point Parameters
456. NeurIPS 2026Unifying Sparsity and Discreteness: One-Shot Pruning for Quantized LLMs via Discrete Optimization
457. NeurIPS 2026Unpaired Canonical Correlation Analysis
458. NeurIPS 2026UpSafe℃: Upcycling for Controllable Safety in Large Language Models
459. NeurIPS 2026Value Entanglement: Conflation Between Different Kinds of Good In (Some) Large Language Models
460. NeurIPS 2026Values Are Not Single Labels: Distributional Value Profiling Across Groups and Contexts
461. NeurIPS 2026Visual Robustness and Neural Alignment in a Shared Foraging Task: The Mouse vs. AI Benchmark
462. NeurIPS 2026Watermarking Without Standards Is Not AI Governance
463. NeurIPS 2026Weird Generalization from Narrow Finetuning: Persona Shifts and Inductive Backdoors
464. NeurIPS 2026Welfare, Improvability, and Variance: A Principal-Agent Approach to Optimal Benchmark Item Aggregation
465. NeurIPS 2026What Can Labels Alone Audit and Edit in Frozen Representations?
466. NeurIPS 2026What Does the Brain See? Multiview Neural Representation to Demystify the Brain-Visual Alignment
467. NeurIPS 2026What Stops SGD on LLM Pre-Training: The Need for Large Learning Rate and How to Achieve it
468. NeurIPS 2026When Alignment Fails: Stabilizing Cross-Dynamics RL with Prototype Trust Regions
469. NeurIPS 2026When and Why SignSGD Outperforms SGD: A Theoretical Study Based on $\ell_1$-norm Lower Bounds
470. NeurIPS 2026When do Prophets Profit in Prediction Markets?
471. NeurIPS 2026When Empathy Misses the Goal: A Benchmark for Goal Displacement in LLM Advice
472. NeurIPS 2026When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models
473. NeurIPS 2026When Must AI Training Stage Checkpoints? A Distributional Model of Durability Boundaries at Scale
474. NeurIPS 2026When Policy Entropy Constraint Fails: Preserving Diversity in Flow-based RLHF via Perceptual Entropy
475. NeurIPS 2026When Safety Becomes An Outlier: Understanding the Retention of LLM Safety Behaviors
476. NeurIPS 2026Which Pairs Should We Compare for DPO?
477. NeurIPS 2026Why Do DiT Editors Drift? Plug-and-Play Low Frequency Alignment in VAE Latent Space
478. NeurIPS 2026Why Invariance is Not Enough for Biomedical Domain Generalization and How to Fix It
479. NeurIPS 2026Words That Make Language Models Perceive
480. NeurIPS 2026XL-SafetyBench: A Country-Grounded Cross-Cultural Benchmark for LLM Safety and Cultural Sensitivity
481. NeurIPS 2025$\Delta \mathrm{Energy}$: Optimizing Energy Change During Vision-Language Alignment Improves both OOD Detection and OOD Generalization
482. NeurIPS 2025$\Psi$-Sampler: Initial Particle Sampling for SMC-Based Inference-Time Reward Alignment in Score Models
483. NeurIPS 2025$O(\sqrt{T})$ Static Regret and Instance Dependent Constraint Violation for Constrained Online Convex Optimization
484. NeurIPS 2025A Closer Look at Graph Transformers: Cross-Aggregation and Beyond
485. NeurIPS 2025A Closer Look at NTK Alignment: Linking Phase Transitions in Deep Image Regression
486. NeurIPS 2025A faster training algorithm for regression trees with linear leaves, and an analysis of its complexity
487. NeurIPS 2025A Gradient Guidance Perspective on Stepwise Preference Optimization for Diffusion Models
488. NeurIPS 2025A Snapshot of Influence: A Local Data Attribution Framework for Online Reinforcement Learning
489. NeurIPS 2025A TRIANGLE Enables Multimodal Alignment Beyond Cosine Similarity
490. NeurIPS 2025AANet: Virtual Screening under Structural Uncertainty via Alignment and Aggregation
491. NeurIPS 2025Accident Anticipation via Temporal Occurrence Prediction
492. NeurIPS 2025Adaptive and Multi-scale Affinity Alignment for Hierarchical Contrastive Learning
493. NeurIPS 2025Adaptive Batch-Wise Sample Scheduling for Direct Preference Optimization
494. NeurIPS 2025Adaptive Defense against Harmful Fine-Tuning for Large Language Models via Bayesian Data Scheduler
495. NeurIPS 2025Adaptive Variance Inflation in Thompson Sampling: Efficiency, Safety, Robustness, and Beyond
496. NeurIPS 2025Adversarial Attacks against Closed-Source MLLMs via Feature Optimal Alignment
497. NeurIPS 2025AdvPrefix: An Objective for Nuanced LLM Jailbreaks
498. NeurIPS 2025AF-UMC: An Alignment-Free Fusion Framework for Unaligned Multi-View Clustering
499. NeurIPS 2025Afterburner: Reinforcement Learning Facilitates Self-Improving Code Efficiency Optimization
500. NeurIPS 2025AgentBreeder: Mitigating the AI Safety Risks of Multi-Agent Scaffolds via Self-Improvement

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。