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 4,558 篇，第 53 / 92 页

  1. ICML 2025Improving LLM Safety Alignment with Dual-Objective Optimization
  2. ICML 2025Improving Model Alignment Through Collective Intelligence of Open-Source Models
  3. ICML 2025Improving Multimodal Learning Balance and Sufficiency through Data Remixing
  4. ICML 2025InfAlign: Inference-aware language model alignment
  5. ICML 2025Inference-Time Alignment of Diffusion Models with Direct Noise Optimization
  6. ICML 2025Invariance Makes LLM Unlearning Resilient Even to Unanticipated Downstream Fine-Tuning
  7. ICML 2025Is Best-of-N the Best of Them? Coverage, Scaling, and Optimality in Inference-Time Alignment
  8. ICML 2025ITFormer: Bridging Time Series and Natural Language for Multi-Modal QA with Large-Scale Multitask Dataset
  9. ICML 2025Jacobian Sparse Autoencoders: Sparsify Computations, Not Just Activations
  10. ICML 2025Joint Metric Space Embedding by Unbalanced Optimal Transport with Gromov–Wasserstein Marginal Penalization
  11. ICML 2025Just Enough Shifts: Mitigating Over-Refusal in Aligned Language Models with Targeted Representation Fine-Tuning
  12. ICML 2025Kernel Quantile Embeddings and Associated Probability Metrics
  13. ICML 2025Knowledge Swapping via Learning and Unlearning
  14. ICML 2025Knowledge-Guided Wasserstein Distributionally Robust Optimization
  15. ICML 2025Language Models as Implicit Tree Search
  16. ICML 2025LapSum - One Method to Differentiate Them All: Ranking, Sorting and Top-k Selection
  17. ICML 2025Larger or Smaller Reward Margins to Select Preferences for LLM Alignment?
  18. ICML 2025Latent Preference Coding: Aligning Large Language Models via Discrete Latent Codes
  19. ICML 2025Layer-wise Alignment: Examining Safety Alignment Across Image Encoder Layers in Vision Language Models
  20. ICML 2025Learning Along the Arrow of Time: Hyperbolic Geometry for Backward-Compatible Representation Learning
  21. ICML 2025Learning Cascade Ranking as One Network
  22. ICML 2025Learning from others' mistakes: Finetuning machine translation models with span-level error annotations
  23. ICML 2025Learning Safety Constraints for Large Language Models
  24. ICML 2025Learning Strategic Language Agents in the Werewolf Game with Iterative Latent Space Policy Optimization
  25. ICML 2025Learning the RoPEs: Better 2D and 3D Position Encodings with STRING
  26. ICML 2025Learnware Specification via Dual Alignment
  27. ICML 2025Leveraging Sparsity for Sample-Efficient Preference Learning: A Theoretical Perspective
  28. ICML 2025Linear Bandits with Partially Observable Features
  29. ICML 2025LIVS: A Pluralistic Alignment Dataset for Inclusive Public Spaces
  30. ICML 2025LLM Alignment as Retriever Optimization: An Information Retrieval Perspective
  31. ICML 2025LLM Data Selection and Utilization via Dynamic Bi-level Optimization
  32. ICML 2025Logarithmic Regret for Online KL-Regularized Reinforcement Learning
  33. ICML 2025LOGO --- Long cOntext aliGnment via efficient preference Optimization
  34. ICML 2025Long-Short Alignment for Effective Long-Context Modeling in LLMs
  35. ICML 2025LoRA-One: One-Step Full Gradient Could Suffice for Fine-Tuning Large Language Models, Provably and Efficiently
  36. ICML 2025M3-JEPA: Multimodal Alignment via Multi-gate MoE based on the Joint-Embedding Predictive Architecture
  37. ICML 2025Measuring Diversity: Axioms and Challenges
  38. ICML 2025Meta-Reinforcement Learning with Adaptation from Human Feedback via Preference-Order-Preserving Task Embedding
  39. ICML 2025MetaOptimize: A Framework for Optimizing Step Sizes and Other Meta-parameters
  40. ICML 2025Mind the Gap: a Spectral Analysis of Rank Collapse and Signal Propagation in Attention Layers
  41. ICML 2025MindAligner: Explicit Brain Functional Alignment for Cross-Subject Visual Decoding from Limited fMRI Data
  42. ICML 2025Mitigating Local Cohesion and Global Sparseness in Graph Contrastive Learning with Fuzzy Boundaries
  43. ICML 2025Mixed-curvature decision trees and random forests
  44. ICML 2025MM-RLHF: The Next Step Forward in Multimodal LLM Alignment
  45. ICML 2025MMedPO: Aligning Medical Vision-Language Models with Clinical-Aware Multimodal Preference Optimization
  46. ICML 2025MONA: Myopic Optimization with Non-myopic Approval Can Mitigate Multi-step Reward Hacking
  47. ICML 2025MPO: An Efficient Post-Processing Framework for Mixing Diverse Preference Alignment
  48. ICML 2025Natural Perturbations for Black-box Training of Neural Networks by Zeroth-Order Optimization
  49. ICML 2025Nesterov Method for Asynchronous Pipeline Parallel Optimization
  50. ICML 2025Neural Collapse Beyond the Unconstrained Features Model: Landscape, Dynamics, and Generalization in the Mean-Field Regime

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