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 12 / 29 页

551. NeurIPS 2025Chain-of-Zoom: Extreme Super-Resolution via Scale Autoregression and Preference Alignment
552. NeurIPS 2025Checklists Are Better Than Reward Models For Aligning Language Models
553. NeurIPS 2025Clean First, Align Later: Benchmarking Preference Data Cleaning for Reliable LLM Alignment
554. NeurIPS 2025Clip-and-Verify: Linear Constraint-Driven Domain Clipping for Accelerating Neural Network Verification
555. NeurIPS 2025Closed-Form Training Dynamics Reveal Learned Features and Linear Structure in Word2Vec-like Models
556. NeurIPS 2025Collective Bargaining in the Information Economy Can Address AI-Driven Power Concentration
557. NeurIPS 2025Combinatorial Ski Rental Problem: Robust and Learning-Augmented Algorithms
558. NeurIPS 2025Combining Cost Constrained Runtime Monitors for AI Safety
559. NeurIPS 2025Comparison requires valid measurement: Rethinking attack success rate comparisons in AI red teaming
560. NeurIPS 2025ComPO: Preference Alignment via Comparison Oracles
561. NeurIPS 2025Composition and Alignment of Diffusion Models using Constrained Learning
562. NeurIPS 2025Comprehensive Assessment and Analysis for NSFW Content Erasure in Text-to-Image Diffusion models
563. NeurIPS 2025Compress Large Language Models via Collaboration Between Learning and Matrix Approximation
564. NeurIPS 2025Computation and Memory-Efficient Model Compression with Gradient Reweighting
565. NeurIPS 2025Concentration and excess risk bounds for imbalanced classification with synthetic oversampling
566. NeurIPS 2025Confidence-Aware With Prototype Alignment for Partial Multi-label Learning
567. NeurIPS 2025Conformal Arbitrage: Risk-Controlled Balancing of Competing Objectives in Language Models
568. NeurIPS 2025Connecting Neural Models Latent Geometries with Relative Geodesic Representations
569. NeurIPS 2025Consistent Supervised-Unsupervised Alignment for Generalized Category Discovery
570. NeurIPS 2025Continuity and Isolation Lead to Doubts or Dilemmas in Large Language Models
571. NeurIPS 2025Contrastive Learning with Data Misalignment: Feature Purity, Training Dynamics and Theoretical Generalization Guarantees
572. NeurIPS 2025CoP: Agentic Red-teaming for Large Language Models using Composition of Principles
573. NeurIPS 2025CoT Red-Handed: Stress Testing Chain-of-Thought Monitoring
574. NeurIPS 2025Counterfactual Reasoning for Steerable Pluralistic Value Alignment of Large Language Models
575. NeurIPS 2025CPO: Condition Preference Optimization for Controllable Image Generation
576. NeurIPS 2025CTRL-ALT-DECEIT Sabotage Evaluations for Automated AI R&D
577. NeurIPS 2025CURE: Concept Unlearning via Orthogonal Representation Editing in Diffusion Models
578. NeurIPS 2025CVGL: Causal Learning and Geometric Topology
579. NeurIPS 2025DAPO : Improving Multi-Step Reasoning Abilities of Large Language Models with Direct Advantage-Based Policy Optimization
580. NeurIPS 2025Decoupling Contrastive Decoding: Robust Hallucination Mitigation in Multimodal Large Language Models
581. NeurIPS 2025Deep Value Benchmark: Measuring Whether Models Generalize Deep values or Shallow Preferences
582. NeurIPS 2025DeepKD: A Deeply Decoupled and Denoised Knowledge Distillation Trainer
583. NeurIPS 2025Delta Attention: Fast and Accurate Sparse Attention Inference by Delta Correction
584. NeurIPS 2025DenseDPO: Fine-Grained Temporal Preference Optimization for Video Diffusion Models
585. NeurIPS 2025Differentiable Sparsity via $D$-Gating: Simple and Versatile Structured Penalization
586. NeurIPS 2025Diffusion Model as a Noise-Aware Latent Reward Model for Step-Level Preference Optimization
587. NeurIPS 2025Diffusion Tree Sampling: Scalable inference‑time alignment of diffusion models
588. NeurIPS 2025Diffusion-Classifier Synergy: Reward-Aligned Learning via Mutual Boosting Loop for FSCIL
589. NeurIPS 2025Dimensional Collapse in VQVAEs: Evidence and Remedies
590. NeurIPS 2025Dimensionality Mismatch Between Brains and Artificial Neural Networks
591. NeurIPS 2025Direct Alignment with Heterogeneous Preferences
592. NeurIPS 2025Directed-Tokens: A Robust Multi-Modality Alignment Approach to Large Language-Vision Models
593. NeurIPS 2025Distillation Robustifies Unlearning
594. NeurIPS 2025Distortion of AI Alignment: Does Preference Optimization Optimize for Preferences?
595. NeurIPS 2025Distributional LLM-as-a-Judge
596. NeurIPS 2025Distributive Fairness in Large Language Models: Evaluating Alignment with Human Values
597. NeurIPS 2025DNAEdit: Direct Noise Alignment for Text-Guided Rectified Flow Editing
598. NeurIPS 2025Do LVLMs Truly Understand Video Anomalies? Revealing Hallucination via Co-Occurrence Patterns
599. NeurIPS 2025Doctor Approved: Generating Medically Accurate Skin Disease Images through AI-Expert Feedback
600. NeurIPS 2025Don’t Forget the Enjoin: FocalLoRA for Instruction Hierarchical Alignment in Large Language Models

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。