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 1,882 篇，第 28 / 38 页

1351. ICML 2026Toward Safe Quantization-Aware Fine-tuning: Understanding and Mitigating Safety Alignment Degradation
1352. ICML 2026Toward Scalable and Valid Conditional Independence Testing with Spectral Representations
1353. ICML 2026Toward Stable Value Alignment: Introducing Independent Modules for Consistent Value Guidance
1354. ICML 2026Towards a Science of AI Agent Reliability
1355. ICML 2026Towards Context-Invariant Safety Alignment for Large Language Models
1356. ICML 2026Towards Disentangled Preference Optimization Dynamics: Suppress the Loser, Preserve the Winner
1357. ICML 2026Towards Multimodal Large Language Models with Both Training and Inference Efficiency
1358. ICML 2026Towards Seed-Robust Safety Alignment in Text-to-Image Models
1359. ICML 2026Towards Streaming Synchronized Spatial Audio Generation via Autoregressive Diffusion Transformer
1360. ICML 2026Towards Understanding Steering Strength
1361. ICML 2026Towards Uniformity and Alignment for Multimodal Representation Learning
1362. ICML 2026TraceRouter: Robust Safety for Large Foundation Models via Path-Level Intervention
1363. ICML 2026Tracing the Dynamics of Refusal: Exploiting Latent Refusal Trajectories for Robust Jailbreak Detection
1364. ICML 2026Training LLM Agents to Empower Humans
1365. ICML 2026Training with Honeypots: Reshaping How LLMs Fail Under Adversarial Attacks
1366. ICML 2026Transitivity Meets Cyclicity: Explicit Preference Decomposition for Dynamic Large Language Model Alignment
1367. ICML 2026Trojan-Speak: Bypassing Constitutional Classifiers with No Jailbreak Tax via Adversarial Finetuning
1368. ICML 2026TUR-DPO: Topology- and Uncertainty-Aware Direct Preference Optimization
1369. ICML 2026Turning Drift into Constraint: Robust Reasoning Alignment in Non-Stationary Multi-Stream Environments
1370. ICML 2026Turning Stale Gradients into Stable Gradients: Coherent Coordinate Descent with Implicit Landscape Smoothing for Lightweight Zeroth-Order Optimization
1371. ICML 2026Unbiased Alignment for Large Language Models with Noisy Preferences
1372. ICML 2026Unbiased Principles, Robust Rewards
1373. ICML 2026Unbiased Reward Modeling from Implicit Feedback for LLM Alignment
1374. ICML 2026Uncovering Hidden Triggers: Backdoor Attribution in Language Models
1375. ICML 2026Understanding and Mitigating Token-Pruning-Induced Vulnerabilities in VLMs
1376. ICML 2026Understanding the Performance Gap in Preference Learning: A Dichotomy of RLHF and DPO
1377. ICML 2026Unifying Adversarial Robustness and Training Across Text Scoring Models
1378. ICML 2026UniRTL: Unifying Code and Graph for Robust RTL Representation Learning
1379. ICML 2026UniScale: Adaptive Unified Inference Scaling via Online Joint Optimization of Model Routing and Test-Time Scaling
1380. ICML 2026Unlearning’s Blind Spots: Over‑Unlearning and Prototypical Relearning Attack
1381. ICML 2026Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents
1382. ICML 2026Unveiling And Addressing Dimensional Collapse In Vector Quantization Models Via Codebook Regularization
1383. ICML 2026VALUEFLOW: Toward Pluralistic and Steerable Value-based Alignment in Large Language Models
1384. ICML 2026VERA-V: Variational Inference Framework for Jailbreaking Vision-Language Models
1385. ICML 2026Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity
1386. ICML 2026ViEEG: Hierarchical Visual Neural Representation for EEG Brain Decoding
1387. ICML 2026VSCD: Video-based Scene Change Detection in Unaligned Scenes
1388. ICML 2026WeightCLIP: Aligning Datasets and Models for Weight Space Learning
1389. ICML 2026When Actions Go Off-Task: Detecting and Correcting Misaligned Actions in Computer-Use Agents
1390. ICML 2026When and How Human Curation Backfires: Preference Alignment under Multi-Model Self-Consuming Loop
1391. ICML 2026When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use Agents
1392. ICML 2026When Distance Distracts: Representation Distance Bias in BT-Loss for Reward Models
1393. ICML 2026When Labelers Stay Silent: The Power of Ties in Cost-Effective Preference Learning
1394. ICML 2026When Preference Labels Fall Short: Aligning Diffusion Models from Real Data
1395. ICML 2026When Search Goes Wrong: Red-Teaming Web-Augmented Large Language Models
1396. ICML 2026When Tabular Foundation Models Meet Strategic Tabular Data: A Prior Alignment Approach
1397. ICML 2026Who Transfers Safety? Identifying and Targeting Cross-Lingual Shared Safety Neurons
1398. ICML 2026Why Deep Jacobian Spectra Separate: Depth-Induced Scaling and Singular-Vector Alignment
1399. ICML 2026Width Independent Bounds for the Local Lipschitz Constant of Deep Neural Networks at Random Initialization and after Lazy Training
1400. ICML 2026WildCat: Near-Linear Attention in Theory and Practice

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。