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
- 社会层面 · safety论文从 2023 年的 4 篇增长到 2026 年的 370 篇，三年间扩大到约 93 倍。
- ICML 是发社会层面 · safety论文最多的会议，2023–2026 年共 364 篇，占四个会议总量的 53%。
- 交叉方向里涨得最快的是对齐与安全：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 3；同组增长最快的社会层面 · privacy是 6.6 倍，社会层面 · safety是 93 倍。
各会议论文数
按年份和会议统计
| 年份 | NeurIPS | ICLR | ICML | CVPR | 合计 |
|---|---|---|---|---|---|
| 2023 | 2 | 0 | 2 | 0 | 4 |
| 2024 | 23 | 11 | 6 | 1 | 41 |
| 2025 | 131 | 11 | 130 | 1 | 273 |
| 2026 | 121 | 20 | 226 | 3 | 370 |
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 2026BarrierSteer: LLM Safety via Learning Barrier Steering
- ICML 2026Anti-Backdoor Coreset Selection via Cumulative Entropy
- ICLR 2026Bi-Criteria Metric Distortion
- CVPR 2026Omni-MMSI Toward Identity-attributed Social Interaction Understanding
- NeurIPS 2025Safety Pretraining: Toward the Next Generation of Safe AI
- ICML 2025$S^2$FGL: Spatial Spectral Federated Graph Learning
- ICLR 2025The Hidden Cost of Waiting for Accurate Predictions
- CVPR 2025Do We Really Need Curated Malicious Data for Safety Alignment in Multi-modal Large Language Models?
- NeurIPS 2024Quantifying the Bitter Lesson: How Safety Benchmarks Measure Capabilities Instead of Safety
- ICML 2024Optimally Improving Cooperative Learning in a Social Setting
- ICLR 2024The Devil is in the Neurons: Interpreting and Mitigating Social Biases in Language Models
- CVPR 2024Digital Life Project: Autonomous 3D Characters with Social Intelligence
- NeurIPS 2023Strategyproof Voting under Correlated Beliefs
- ICML 2023Adaptive Coordination in Social Embodied Rearrangement
- NeurIPS 2026Beyond Feature Disruption: Boundary-Diverting Unlearnable Examples against Linear Probing
- ICML 2026AutoControl Arena: Synthesizing Executable Test Environments for Frontier AI Risk Evaluation
- ICLR 2026PRISON: Unmasking the Criminal Potential of Large Language Models
- CVPR 2026Spot The Ball A Benchmark for Visual Social Inference
- NeurIPS 2025SAFEx: Analyzing Vulnerabilities of MoE-Based LLMs via Stable Safety-critical Expert Identification
- ICML 2025Adversaries Can Misuse Combinations of Safe Models
全部社会层面 · safety论文
共 364 篇，第 5 / 8 页
- ICML 2026Through the Stealth Lens: Attention-Aware Defenses Against Poisoning in RAG
- ICML 2026Token-Efficient Change Detection in LLM APIs
- ICML 2026Toward Calibrated Mixture-of-Experts Under Distribution Shift
- ICML 2026Toward Stable Value Alignment: Introducing Independent Modules for Consistent Value Guidance
- ICML 2026Towards Context-Invariant Safety Alignment for Large Language Models
- ICML 2026Towards Functional Correctness of Large Code Models with Selective Generation
- ICML 2026Towards Seed-Robust Safety Alignment in Text-to-Image Models
- ICML 2026Towards Steering without Sacrifice: Principled Training of Steering Vectors for Prompt-only Interventions
- ICML 2026TraceRouter: Robust Safety for Large Foundation Models via Path-Level Intervention
- ICML 2026TSFAdv: Frequency-Guided Black-Box Adversarial Attacks on Time Series Forecasting
- ICML 2026Understanding and Mitigating Token-Pruning-Induced Vulnerabilities in VLMs
- ICML 2026Unified Safe In-context Image Generation in Multimodal Diffusion Transformers via Restricting Unsafe Information Flows
- ICML 2026Unlearning Isn't Deletion: Investigating Reversibility of Machine Unlearning in LLMs
- ICML 2026Unsafer in Many Turns: Benchmarking and Defending Multi-Turn Safety Risks in Tool-Using Agents
- ICML 2026Weight-sparse transformers have interpretable circuits
- ICML 2026What is Missing? Explaining Neurons Activated by Absent Concepts
- ICML 2026When Benign Inputs Lead to Severe Harms: Eliciting Unsafe Unintended Behaviors of Computer-Use Agents
- ICML 2026When Embedding-Based Defenses Fail: Rethinking Safety in LLM-Based Multi-Agent Systems
- ICML 2026When Random Saliency Looks Trained: Architectural Center Bias in CNN Interpretability
- ICML 2026When the Prompt Becomes Visual: Vision-Centric Jailbreak Attacks for Large Image Editing Models
- ICML 2026Who Evaluates AI's Social Impacts? Mapping Coverage and Gaps in First and Third Party Evaluations
- ICML 2026Who Gets Credit or Blame? Attributing Accountability in Modern AI Systems
- ICML 2026Who Transfers Safety? Identifying and Targeting Cross-Lingual Shared Safety Neurons
- ICML 2026X-EviProbe: Post-hoc Parameter-Free Evidential Uncertainty Quantification for Frozen Graph Neural Networks
- ICML 2026Z-Erase: Enabling Concept Erasure in Single Stream Diffusion Transformers
- ICML 2026ZeroUnlearn: Few-Shot Knowledge Unlearning in Large Language Models
- ICML 2025(How) Can Transformers Predict Pseudo-Random Numbers?
- ICML 2025$S^2$FGL: Spatial Spectral Federated Graph Learning
- ICML 2025A Checks-and-Balances Framework for Context-Aware Ethical AI Alignment
- ICML 2025A Lens into Interpretable Transformer Mistakes via Semantic Dependency
- ICML 2025A Non-Asymptotic Convergent Analysis for Scored-Based Graph Generative Model via a System of Stochastic Differential Equations
- ICML 2025A Versatile Influence Function for Data Attribution with Non-Decomposable Loss
- ICML 2025Activation Space Interventions Can Be Transferred Between Large Language Models
- ICML 2025Adversarial Reasoning at Jailbreaking Time
- ICML 2025Adversaries Can Misuse Combinations of Safe Models
- ICML 2025Algorithmic Recourse for Long-Term Improvement
- ICML 2025An Efficient Private GPT Never Autoregressively Decodes
- ICML 2025An End-to-End Model for Logits-Based Large Language Models Watermarking
- ICML 2025Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning Attack
- ICML 2025Assessing Safety Risks and Quantization-aware Safety Patching for Quantized Large Language Models
- ICML 2025Automated Red Teaming with GOAT: the Generative Offensive Agent Tester
- ICML 2025Automatically Interpreting Millions of Features in Large Language Models
- ICML 2025Avoiding Leakage Poisoning: Concept Interventions Under Distribution Shifts
- ICML 2025AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders
- ICML 2025B-score: Detecting biases in large language models using response history
- ICML 2025CASE-Bench: Context-Aware SafEty Benchmark for Large Language Models
- ICML 2025Certified Unlearning for Neural Networks
- ICML 2025Circumventing Backdoor Space via Weight Symmetry
- ICML 2025Confidential Guardian: Cryptographically Prohibiting the Abuse of Model Abstention
- ICML 2025Correlated Errors in Large Language Models
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"}
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。