社会层面 · safety论文趋势 2023–2026

AI 社会议题：safety

统计范围：NeurIPS、ICLR、ICML、CVPR 四个会议，2023–2026 年，共 688 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

社会层面 · safety论文数，按年份和会议
年份NeurIPSICLRICMLCVPR合计
202320204
202423116141
2025131111301273
2026121202263370

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

全部社会层面 · safety论文

年份全部2023202420252026
会议全部NeurIPSICLRICMLCVPR

共 277 篇，第 5 / 6 页

  1. NeurIPS 2025Representational Difference Explanations
  2. NeurIPS 2025Residual Stream Analysis of Overfitting And Structural Disruptions
  3. NeurIPS 2025RiOSWorld: Benchmarking the Risk of Multimodal Computer-Use Agents
  4. NeurIPS 2025Risk Management for Mitigating Benchmark Failure Modes: BenchRisk
  5. NeurIPS 2025Robust Explanations of Graph Neural Networks via Graph Curvatures
  6. NeurIPS 2025Safe RLHF-V: Safe Reinforcement Learning from Multi-modal Human Feedback
  7. NeurIPS 2025SAFEPATH: Preventing Harmful Reasoning in Chain-of-Thought via Early Alignment
  8. NeurIPS 2025Safety Depth in Large Language Models: A Markov Chain Perspective
  9. NeurIPS 2025Safety Pretraining: Toward the Next Generation of Safe AI
  10. NeurIPS 2025SafeVid: Toward Safety Aligned Video Large Multimodal Models
  11. NeurIPS 2025SAFEx: Analyzing Vulnerabilities of MoE-Based LLMs via Stable Safety-critical Expert Identification
  12. NeurIPS 2025SAGE-Eval: Evaluating LLMs for Systematic Generalizations of Safety Facts
  13. NeurIPS 2025Sample-efficient Learning of Concepts with Theoretical Guarantees: from Data to Concepts without Interventions
  14. NeurIPS 2025Scalable Fingerprinting of Large Language Models
  15. NeurIPS 2025Scalable Valuation of Human Feedback through Provably Robust Model Alignment
  16. NeurIPS 2025Scaling Laws For Scalable Oversight
  17. NeurIPS 2025Semantic Representation Attack against Aligned Large Language Models
  18. NeurIPS 2025SHAP Meets Tensor Networks: Provably Tractable Explanations with Parallelism
  19. NeurIPS 2025SHAP values via sparse Fourier representation
  20. NeurIPS 2025SHAP zero Explains Biological Sequence Models with Near-zero Marginal Cost for Future Queries
  21. NeurIPS 2025Shape it Up! Restoring LLM Safety during Finetuning
  22. NeurIPS 2025Smoothed Differentiation Efficiently Mitigates Shattered Gradients in Explanations
  23. NeurIPS 2025SpEx: A Spectral Approach to Explainable Clustering
  24. NeurIPS 2025Spurious-Aware Prototype Refinement for Reliable Out-of-Distribution Detection
  25. NeurIPS 2025STEAD: Robust Provably Secure Linguistic Steganography with Diffusion Language Model
  26. NeurIPS 2025StegoZip: Enhancing Linguistic Steganography Payload in Practice with Large Language Models
  27. NeurIPS 2025Strategic Classification with Non-Linear Classifiers
  28. NeurIPS 2025Strategic Cost Selection in Participatory Budgeting
  29. NeurIPS 2025Strategic Hypothesis Testing
  30. NeurIPS 2025TAI3: Testing Agent Integrity in Interpreting User Intent
  31. NeurIPS 2025Taming Hyperparameter Sensitivity in Data Attribution: Practical Selection Without Costly Retraining
  32. NeurIPS 2025Taught Well Learned Ill: Towards Distillation-conditional Backdoor Attack
  33. NeurIPS 2025The Future Unmarked: Watermark Removal in AI-Generated Images via Next-Frame Prediction
  34. NeurIPS 2025The VLLM Safety Paradox: Dual Ease in Jailbreak Attack and Defense
  35. NeurIPS 2025Towards Resilient Safety-driven Unlearning for Diffusion Models against Downstream Fine-tuning
  36. NeurIPS 2025Towards Understanding Safety Alignment: A Mechanistic Perspective from Safety Neurons
  37. NeurIPS 2025Transformer Key-Value Memories Are Nearly as Interpretable as Sparse Autoencoders
  38. NeurIPS 2025Uncovering a Universal Abstract Algorithm for Modular Addition in Neural Networks
  39. NeurIPS 2025Unlearning-Aware Minimization
  40. NeurIPS 2025Unmasking Puppeteers: Leveraging Biometric Leakage to Expose Impersonation in AI-Based Videoconferencing
  41. NeurIPS 2025Unveiling Environmental Sensitivity of Individual Gains in Influence Maximization
  42. NeurIPS 2025Vector Database Watermarking
  43. NeurIPS 2025Versatile Transferable Unlearnable Example Generator
  44. NeurIPS 2025Video-SafetyBench: A Benchmark for Safety Evaluation of Video LVLMs
  45. NeurIPS 2025Virus Infection Attack on LLMs: Your Poisoning Can Spread "VIA" Synthetic Data
  46. NeurIPS 2025VLMs can Aggregate Scattered Training Patches
  47. NeurIPS 2025What is Your Data Worth to GPT? LLM-Scale Data Valuation with Influence Functions
  48. NeurIPS 2025What Really is a Member? Discrediting Membership Inference via Poisoning
  49. NeurIPS 2025Whose View of Safety? A Deep DIVE Dataset for Pluralistic Alignment of Text-to-Image Models
  50. NeurIPS 2025Why Do Some Language Models Fake Alignment While Others Don't?

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 查社会层面 · safety在 ICML 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": "social-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社会层面 · safety论文？
共 688 篇，按年份分别是 2023 年 4 篇、2024 年 41 篇、2025 年 273 篇、2026 年 370 篇。
NeurIPS 2026 有多少篇社会层面 · safety论文？
NeurIPS 2026 有 121 篇社会层面 · safety论文，上一届是 131 篇。
ICLR 2026 有多少篇社会层面 · safety论文？
ICLR 2026 有 20 篇社会层面 · safety论文，上一届是 11 篇。
ICML 2026 有多少篇社会层面 · safety论文？
ICML 2026 有 226 篇社会层面 · safety论文，上一届是 130 篇。
CVPR 2026 有多少篇社会层面 · safety论文？
CVPR 2026 有 3 篇社会层面 · safety论文，上一届是 1 篇。
哪个会议的社会层面 · safety论文最多？
ICML，2023–2026 年共 364 篇。

「对齐与负责任 AI」下的其他主题