对齐与安全论文趋势 2023–2026

RLHF、偏好优化、红队与安全对齐

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 4,558 篇被归入该主题的论文。

主要结论

每年论文数

各会议论文数

按年份和会议统计

对齐与安全论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
202313887101112438
2024348156246125875
20254554733001351,363
20264806235512281,882

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

全部对齐与安全论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 1,339 篇，第 27 / 27 页

  1. ICLR 2023One-Pixel Shortcut: On the Learning Preference of Deep Neural Networks
  2. ICLR 2023Ordered GNN: Ordering Message Passing to Deal with Heterophily and Over-smoothing
  3. ICLR 2023Partial Label Unsupervised Domain Adaptation with Class-Prototype Alignment
  4. ICLR 2023PASHA: Efficient HPO and NAS with Progressive Resource Allocation
  5. ICLR 2023Personalized Federated Learning with Feature Alignment and Classifier Collaboration
  6. ICLR 2023PiFold: Toward effective and efficient protein inverse folding
  7. ICLR 2023Quantifying and Mitigating the Impact of Label Errors on Model Disparity Metrics
  8. ICLR 2023Relative Behavioral Attributes: Filling the Gap between Symbolic Goal Specification and Reward Learning from Human Preferences
  9. ICLR 2023Reliability of CKA as a Similarity Measure in Deep Learning
  10. ICLR 2023Restricted Strong Convexity of Deep Learning Models with Smooth Activations
  11. ICLR 2023Safe Exploration Incurs Nearly No Additional Sample Complexity for Reward-Free RL
  12. ICLR 2023Safe Reinforcement Learning From Pixels Using a Stochastic Latent Representation
  13. ICLR 2023Scalable and Equivariant Spherical CNNs by Discrete-Continuous (DISCO) Convolutions
  14. ICLR 2023Semi-Parametric Inducing Point Networks and Neural Processes
  15. ICLR 2023Simple and Scalable Nearest Neighbor Machine Translation
  16. ICLR 2023Simple initialization and parametrization of sinusoidal networks via their kernel bandwidth
  17. ICLR 2023Simplicial Embeddings in Self-Supervised Learning and Downstream Classification
  18. ICLR 2023Sparse Mixture-of-Experts are Domain Generalizable Learners
  19. ICLR 2023Stochastic Differentially Private and Fair Learning
  20. ICLR 2023Supervision Complexity and its Role in Knowledge Distillation
  21. ICLR 2023Symmetries, Flat Minima, and the Conserved Quantities of Gradient Flow
  22. ICLR 2023SYNC: SAFETY-AWARE NEURAL CONTROL FOR STABILIZING STOCHASTIC DELAY-DIFFERENTIAL EQUATIONS
  23. ICLR 2023TANGOS: Regularizing Tabular Neural Networks through Gradient Orthogonalization and Specialization
  24. ICLR 2023Targeted Hyperparameter Optimization with Lexicographic Preferences Over Multiple Objectives
  25. ICLR 2023Task Ambiguity in Humans and Language Models
  26. ICLR 2023TempCLR: Temporal Alignment Representation with Contrastive Learning
  27. ICLR 2023Thalamus: a brain-inspired algorithm for biologically-plausible continual learning and disentangled representations
  28. ICLR 2023Towards One-shot Neural Combinatorial Solvers: Theoretical and Empirical Notes on the Cardinality-Constrained Case
  29. ICLR 2023Towards Understanding and Mitigating Dimensional Collapse in Heterogeneous Federated Learning
  30. ICLR 2023Towards Understanding Ensemble, Knowledge Distillation and Self-Distillation in Deep Learning
  31. ICLR 2023Training language models to summarize narratives improves brain alignment
  32. ICLR 2023Understanding Influence Functions and Datamodels via Harmonic Analysis
  33. ICLR 2023Unified Detoxifying and Debiasing in Language Generation via Inference-time Adaptive Optimization
  34. ICLR 2023Unsupervised Manifold Alignment with Joint Multidimensional Scaling
  35. ICLR 2023Voxurf: Voxel-based Efficient and Accurate Neural Surface Reconstruction
  36. ICLR 2023What Is Missing in IRM Training and Evaluation? Challenges and Solutions
  37. ICLR 2023When to Make and Break Commitments?
  38. ICLR 2023Where to Begin? On the Impact of Pre-Training and Initialization in Federated Learning
  39. ICLR 2023Why (and When) does Local SGD Generalize Better than SGD?

在 Claude、Cursor 里直接问这个趋势

接上 Paperlist MCP 后，写综述、挑方向或准备投稿时，不用离开对话框就能拿到这页的数字：每年各会议的论文数、增长倍数、涨得最快的子方向和代表论文，AI 会按你的问题整理。

可以直接复制这句话问 AI

用 paperlist 查对齐与安全在 NeurIPS 2023–2026 的论文趋势，哪一年涨得最快？

AI 会调用的工具：get_topic_stats {"topic": "alignment-and-safety"}

两分钟接入 Paperlist MCP（Claude Code、Cursor、Codex、ChatGPT）

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。

「对齐与负责任 AI」下的其他主题