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 1,363 篇，第 8 / 28 页

351. ICLR 2025Interpreting Language Reward Models via Contrastive Explanations
352. ICLR 2025Inverse Constitutional AI: Compressing Preferences into Principles
353. ICLR 2025Is In-Context Learning Sufficient for Instruction Following in LLMs?
354. ICLR 2025Is Your Multimodal Language Model Oversensitive to Safe Queries?
355. ICLR 2025Iterative Label Refinement Matters More than Preference Optimization under Weak Supervision
356. ICLR 2025Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning
357. ICLR 2025Jailbreak Antidote: Runtime Safety-Utility Balance via Sparse Representation Adjustment in Large Language Models
358. ICLR 2025Jailbreaking as a Reward Misspecification Problem
359. ICLR 2025Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks
360. ICLR 2025Joint Graph Rewiring and Feature Denoising via Spectral Resonance
361. ICLR 2025Joint Reward and Policy Learning with Demonstrations and Human Feedback Improves Alignment
362. ICLR 2025Judge Decoding: Faster Speculative Sampling Requires Going Beyond Model Alignment
363. ICLR 2025JudgeLM: Fine-tuned Large Language Models are Scalable Judges
364. ICLR 2025Justice or Prejudice? Quantifying Biases in LLM-as-a-Judge
365. ICLR 2025L3Ms — Lagrange Large Language Models
366. ICLR 2025Language Model Alignment in Multilingual Trolley Problems
367. ICLR 2025Language Models are Advanced Anonymizers
368. ICLR 2025Language Models Learn to Mislead Humans via RLHF
369. ICLR 2025Large Language Models can Become Strong Self-Detoxifiers
370. ICLR 2025Large-scale and Fine-grained Vision-language Pre-training for Enhanced CT Image Understanding
371. ICLR 2025Latent Safety-Constrained Policy Approach for Safe Offline Reinforcement Learning
372. ICLR 2025Latent Space Chain-of-Embedding Enables Output-free LLM Self-Evaluation
373. ICLR 2025LeanQuant: Accurate and Scalable Large Language Model Quantization with Loss-error-aware Grid
374. ICLR 2025Learn Your Reference Model for Real Good Alignment
375. ICLR 2025Learning and aligning single-neuron invariance manifolds in visual cortex
376. ICLR 2025Learning Causal Alignment for Reliable Disease Diagnosis
377. ICLR 2025Learning Diverse Attacks on Large Language Models for Robust Red-Teaming and Safety Tuning
378. ICLR 2025Learning Dynamics of LLM Finetuning
379. ICLR 2025Learning from End User Data with Shuffled Differential Privacy over Kernel Densities
380. ICLR 2025Learning from negative feedback, or positive feedback or both
381. ICLR 2025Learning High-Degree Parities: The Crucial Role of the Initialization
382. ICLR 2025Learning LLM-as-a-Judge for Preference Alignment
383. ICLR 2025Learning to Clarify: Multi-turn Conversations with Action-Based Contrastive Self-Training
384. ICLR 2025Learning to Explore and Exploit with GNNs for Unsupervised Combinatorial Optimization
385. ICLR 2025LiNeS: Post-training Layer Scaling Prevents Forgetting and Enhances Model Merging
386. ICLR 2025LLaMaFlex: Many-in-one LLMs via Generalized Pruning and Weight Sharing
387. ICLR 2025LLaVA-MoD: Making LLaVA Tiny via MoE-Knowledge Distillation
388. ICLR 2025LLM Unlearning via Loss Adjustment with Only Forget Data
389. ICLR 2025Local Patterns Generalize Better for Novel Anomalies
390. ICLR 2025Locality Alignment Improves Vision-Language Models
391. ICLR 2025Logicbreaks: A Framework for Understanding Subversion of Rule-based Inference
392. ICLR 2025Long-tailed Adversarial Training with Self-Distillation
393. ICLR 2025LongPO: Long Context Self-Evolution of Large Language Models through Short-to-Long Preference Optimization
394. ICLR 2025Looking Inward: Language Models Can Learn About Themselves by Introspection
395. ICLR 2025MA-RLHF: Reinforcement Learning from Human Feedback with Macro Actions
396. ICLR 2025Machine Unlearning Fails to Remove Data Poisoning Attacks
397. ICLR 2025MACPO: Weak-to-Strong Alignment via Multi-Agent Contrastive Preference Optimization
398. ICLR 2025Magnetic Preference Optimization: Achieving Last-iterate Convergence for Language Model Alignment
399. ICLR 2025Magpie: Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing
400. ICLR 2025MallowsPO: Fine-Tune Your LLM with Preference Dispersions

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。