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 246 篇，第 4 / 5 页

  1. ICML 2024Perfect Alignment May be Poisonous to Graph Contrastive Learning
  2. ICML 2024Perturb-and-Project: Differentially Private Similarities and Marginals
  3. ICML 2024Piecewise Constant and Linear Regression Trees: An Optimal Dynamic Programming Approach
  4. ICML 2024PlanDQ: Hierarchical Plan Orchestration via D-Conductor and Q-Performer
  5. ICML 2024Position: A Roadmap to Pluralistic Alignment
  6. ICML 2024Position: A Safe Harbor for AI Evaluation and Red Teaming
  7. ICML 2024Position: Beyond Personhood: Agency, Accountability, and the Limits of Anthropomorphic Ethical Analysis
  8. ICML 2024Position: Evolving AI Collectives Enhance Human Diversity and Enable Self-Regulation
  9. ICML 2024Position: Explain to Question not to Justify
  10. ICML 2024Position: Intent-aligned AI Systems Must Optimize for Agency Preservation
  11. ICML 2024Position: Near to Mid-term Risks and Opportunities of Open-Source Generative AI
  12. ICML 2024Position: Technical Research and Talent is Needed for Effective AI Governance
  13. ICML 2024Position: The No Free Lunch Theorem, Kolmogorov Complexity, and the Role of Inductive Biases in Machine Learning
  14. ICML 2024Position: Towards Unified Alignment Between Agents, Humans, and Environment
  15. ICML 2024Position: TrustLLM: Trustworthiness in Large Language Models
  16. ICML 2024Positive and Unlabeled Learning with Controlled Probability Boundary Fence
  17. ICML 2024Preference Optimization for Molecule Synthesis with Conditional Residual Energy-based Models
  18. ICML 2024Pricing with Contextual Elasticity and Heteroscedastic Valuation
  19. ICML 2024Principled Penalty-based Methods for Bilevel Reinforcement Learning and RLHF
  20. ICML 2024Private and Federated Stochastic Convex Optimization: Efficient Strategies for Centralized Systems
  21. ICML 2024Probabilistic Inference in Language Models via Twisted Sequential Monte Carlo
  22. ICML 2024Promoting External and Internal Equities Under Ex-Ante/Ex-Post Metrics in Online Resource Allocation
  23. ICML 2024Prompting4Debugging: Red-Teaming Text-to-Image Diffusion Models by Finding Problematic Prompts
  24. ICML 2024Provably Robust DPO: Aligning Language Models with Noisy Feedback
  25. ICML 2024Quality Diversity through Human Feedback: Towards Open-Ended Diversity-Driven Optimization
  26. ICML 2024Quality-Diversity with Limited Resources
  27. ICML 2024Quantum Algorithms and Lower Bounds for Finite-Sum Optimization
  28. ICML 2024QuIP$\#$: Even Better LLM Quantization with Hadamard Incoherence and Lattice Codebooks
  29. ICML 2024Random Exploration in Bayesian Optimization: Order-Optimal Regret and Computational Efficiency
  30. ICML 2024Random Scaling and Momentum for Non-smooth Non-convex Optimization
  31. ICML 2024Reducing Fine-Tuning Memory Overhead by Approximate and Memory-Sharing Backpropagation
  32. ICML 2024Reducing Item Discrepancy via Differentially Private Robust Embedding Alignment for Privacy-Preserving Cross Domain Recommendation
  33. ICML 2024ReMax: A Simple, Effective, and Efficient Reinforcement Learning Method for Aligning Large Language Models
  34. ICML 2024Representation Surgery: Theory and Practice of Affine Steering
  35. ICML 2024Residual-Conditioned Optimal Transport: Towards Structure-Preserving Unpaired and Paired Image Restoration
  36. ICML 2024Rethinking Optimization and Architecture for Tiny Language Models
  37. ICML 2024Reward Model Learning vs. Direct Policy Optimization: A Comparative Analysis of Learning from Human Preferences
  38. ICML 2024Rewards-in-Context: Multi-objective Alignment of Foundation Models with Dynamic Preference Adjustment
  39. ICML 2024Risk Estimation in a Markov Cost Process: Lower and Upper Bounds
  40. ICML 2024RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback
  41. ICML 2024RLVF: Learning from Verbal Feedback without Overgeneralization
  42. ICML 2024Robust Stable Spiking Neural Networks
  43. ICML 2024Robustly Learning Single-Index Models via Alignment Sharpness
  44. ICML 2024Roping in Uncertainty: Robustness and Regularization in Markov Games
  45. ICML 2024Safe and Robust Subgame Exploitation in Imperfect Information Games
  46. ICML 2024Safe Exploration in Dose Finding Clinical Trials with Heterogeneous Participants
  47. ICML 2024Sarah Frank-Wolfe: Methods for Constrained Optimization with Best Rates and Practical Features
  48. ICML 2024Scalable AI Safety via Doubly-Efficient Debate
  49. ICML 2024Scaling Exponents Across Parameterizations and Optimizers
  50. ICML 2024Self-Alignment of Large Language Models via Monopolylogue-based Social Scene Simulation

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