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 10 / 27 页

451. ICLR 2026Rethinking LoRA for Privacy-Preserving Federated Learning in Large Models
452. ICLR 2026Rethinking Pareto Frontier: On the Optimal Trade-offs in Fair Classification
453. ICLR 2026Rethinking Radiology Report Generation: From Narrative Flow to Topic-Guided Findings
454. ICLR 2026Rethinking the Gold Standard: Why Discrete Curvature Fails to Fully Capture Over-squashing in GNNs?
455. ICLR 2026Reward Model Routing in Alignment
456. ICLR 2026Reward Models Inherit Value Biases from Pretraining
457. ICLR 2026RewardEval: Advancing Reward Model Evaluation
458. ICLR 2026Riemannian Optimization on Relaxed Indicator Matrix Manifold
459. ICLR 2026Risk Phase Transitions in Spiked Regression: Alignment Driven Benign and Catastrophic Overfitting
460. ICLR 2026Risk-Sensitive Agent Compositions
461. ICLR 2026RLBFF: Binary Flexible Feedback to bridge between Human Feedback & Verifiable Rewards
462. ICLR 2026Robust Adversarial Attacks Against Unknown Disturbance via Inverse Gradient Sample
463. ICLR 2026Robust LLM Unlearning via Post Judgment and Multi-round Thinking
464. ICLR 2026Robust Multi-Objective Controlled Decoding of Large Language Models
465. ICLR 2026Robust Preference Alignment via Directional Neighborhood Consensus
466. ICLR 2026Robust Preference Optimization: Aligning Language Models with Noisy Preference Feedback
467. ICLR 2026Robust Reward Modeling via Causal Rubrics
468. ICLR 2026Routing Manifold Alignment Improves Generalization of Mixture-of-Experts LLMs
469. ICLR 2026Routing, Cascades, and User Choice for LLMs
470. ICLR 2026Rubrics as Rewards: Reinforcement Learning Beyond Verifiable Domains
471. ICLR 2026SafeDialBench: A Fine-Grained Safety Evaluation Benchmark for Large Language Models in Multi-Turn Dialogues with Diverse Jailbreak Attacks
472. ICLR 2026SafeDPO: A Simple Approach to Direct Preference Optimization with Enhanced Safety
473. ICLR 2026SafeMoE: Safe Fine-Tuning for MoE LLMs by Aligning Harmful Input Routing
474. ICLR 2026Safety at One Shot: Patching Fine-Tuned LLMs with A Single Instance
475. ICLR 2026Safety Instincts: LLMs Learn to Trust Their Internal Compass for Self-Defense
476. ICLR 2026Safety Mirage: How Spurious Correlations Undermine VLM Safety Fine-Tuning and Can Be Mitigated by Machine Unlearning
477. ICLR 2026Safety Subspaces are Not Linearly Distinct: A Fine-Tuning Case Study
478. ICLR 2026SAGA: Structural Aggregation Guided Alignment with Dynamic View and Neighborhood Order Selection for Multiview Graph Domain Adaptation
479. ICLR 2026Sample Reward Soups: Query-efficient Multi-Reward Guidance for Text-to-Image Diffusion Models
480. ICLR 2026Sampling-aware Adversarial Attacks Against Large Language Models
481. ICLR 2026SDErasure: Concept-Specific Trajectory Shifting for Concept Erasure via Adaptive Diffusion Classifier
482. ICLR 2026SecP-Tuning: Efficient Privacy-Preserving Prompt Tuning for Large Language Models via MPC
483. ICLR 2026Secret-Protected Evolution for Differentially Private Synthetic Text Generation
484. ICLR 2026Secure Outlier-Aware Large Language Model Inference
485. ICLR 2026SEED-SET: Scalable Evolving Experimental Design for System-level Ethical Testing
486. ICLR 2026SeedPrints: Fingerprints Can Even Tell Which Seed Your Large Language Model Was Trained From
487. ICLR 2026Seeing What’s Wrong: A Trajectory-Guided Approach to Caption Error Detection
488. ICLR 2026Selective Data Removal for Distributional Machine Unlearning
489. ICLR 2026Self-Destructive Language Models
490. ICLR 2026Self-Jailbreaking: Language Models Can Reason Themselves Out of Safety Alignment After Benign Reasoning Training
491. ICLR 2026SEMA: Simple yet Effective Learning for Multi-Turn Jailbreak Attacks
492. ICLR 2026Semantic-aware Wasserstein Policy Regularization for Large Language Model Alignment
493. ICLR 2026Semi-Supervised Preference Optimization with Limited Feedback
494. ICLR 2026SeRI: Gradient-Free Sensitive Region Identification in Decision-Based Black-Box Attacks
495. ICLR 2026Set Representation Auxiliary Learning with Adversarial Encoding Perturbation and Optimization
496. ICLR 2026Sharpness-Aware Machine Unlearning
497. ICLR 2026Sharpness-Aware Minimization in Logit Space Efficiently Enhances Direct Preference Optimization
498. ICLR 2026SIGMark: Scalable In-Generation Watermark with Blind Extraction for Video Diffusion
499. ICLR 2026Silent Leaks: Implicit Knowledge Extraction Attack on RAG Systems
500. ICLR 2026SimBench: Benchmarking the Ability of Large Language Models to Simulate Human Behaviors

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。