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 370 篇，第 4 / 8 页
- ICML 2026OpenDeception: Learning Deception and Trust in Human–AI Interaction via Multi-Agent Simulation
- ICML 2026Optimal Regularization for Performative Learning
- ICML 2026Optimal Transport Group Counterfactual Explanations
- ICML 2026Orthogonal Concept Erasure for Diffusion Models
- ICML 2026OSCS: Online Selection with Provable FAR Control for LLM Safety
- ICML 2026Peer-Preservation in Frontier Models
- ICML 2026PersistBench: When Should Long-Term Memories Be Forgotten by LLMs?
- ICML 2026PlugGuard: A Streaming Safeguard for Large Models via Latent Dynamics-Guided Risk Detection
- ICML 2026Pluralistic Leaderboards
- ICML 2026Poison with Style: A Practical Poisoning Attack on Code Large Language Models
- ICML 2026Policies Permitting LLM Use for Polishing Peer Reviews Are Currently Not Enforceable
- ICML 2026Pressure Reveals Character: Behavioural Alignment Evaluation at Depth
- ICML 2026Proactive Defense Benchmark against Deepfake Generation
- ICML 2026Probabilistic Modeling of Latent Agentic Substructures in Deep Neural Networks
- ICML 2026Profiling the Irrational Agent: Cognitive Modeling of LLM Behaviors in Sequential Jailbreaks
- ICML 2026Prompt Injection as Role Confusion
- ICML 2026Quantifying LLM Attention-Head Stability: Implications for Circuit Universality
- ICML 2026Quantifying the Salience of Geo-Cultural Values for Pluralistic Safety Alignment
- ICML 2026Query Lens: Interpreting Sparse Key-Value Features with Indirect Effects
- ICML 2026Questioning the Coverage-Length Metric in Conformal Prediction: When Shorter Intervals Are Not Better
- ICML 2026REALISTA: Realistic Latent Adversarial Attacks that Elicit LLM Hallucinations
- ICML 2026Reasoning Models Struggle to Control their Chains of Thought
- ICML 2026RedDebate: Safer Responses Through Multi-Agent Red Teaming Debates
- ICML 2026Reflector: Internalizing Step-wise Reflection against Indirect Jailbreaks
- ICML 2026Reverse-Engineering Model Editing on Language Models
- ICML 2026Revisiting Robustness for LLM Safety Alignment via Selective Geometry Control
- ICML 2026Reward Hacking Benchmark: Measuring Exploits in LLM Agents with Tool Use
- ICML 2026Risk Awareness Injection: Calibrating Vision-Language Models for Safety without Compromising Utility
- ICML 2026Robust AI Evaluation through Maximal Lotteries
- ICML 2026Robust Human-AI Complementarity under Uncertainty
- ICML 2026Rotation-Invariant Spherical Watermarking via Third-Order SO(3) Representation Coupling
- ICML 2026SAEs-BrainMap: Unveiling the Emergence of Specialized Concepts in Deep Models via Brain Alignment
- ICML 2026Safe Autoregressive Image Generation with Iterative Self-Improving Codebooks
- ICML 2026SafeSearch: Automated Red-Teaming of LLM-Based Search Agents
- ICML 2026Safety Generalization Under Distribution Shift in Safe Reinforcement Learning: A Diabetes Testbed
- ICML 2026Safety-Efficacy Trade Off: Robustness against Data-Poisoning
- ICML 2026Same Question, Different Lies: Cross-Context Consistency (C³) for Black-Box Sandbagging Detection
- ICML 2026SARSteer: Safeguarding Large Audio Language Models via Safe-Ablated Refusal Steering
- ICML 2026SecCodePRM: A Process Reward Model for Code Security
- ICML 2026Securing Multi-Agent Systems Against Corruptions via Node Contribution Backpropagation
- ICML 2026Securing Multimodal AI through Internal Information Decomposition
- ICML 2026Social Hippocampus Memory Learning
- ICML 2026SOPE: Situation-Aware and Statistically Indistinguishable Privacy Exfiltration for MCP-enabled Agents
- ICML 2026Sparse and Faithful Local Explanations with Piecewise Linear Surrogates
- ICML 2026Sparse but Wrong: Incorrect L0 Leads to Incorrect Features in Sparse Autoencoders
- ICML 2026Sparse Models, Sparse Safety: Unsafe Routes in Mixture-of-Experts LLMs
- ICML 2026Sparse Tokens Suffice: Jailbreaking Audio Language Models via Token-Aware Gradient Optimization
- ICML 2026SpatialJB: How Text Distribution Art Becomes The "Jailbreak Key" for LLM Guardrails
- ICML 2026Speculative Safety Honeypot: Toward Proactive Defense Against Multi-turn Agent Attacks
- ICML 2026Speech-Audio Compositional Attacks on Multimodal LLMs and Their Defense with SALMONN-Guard
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。