社会层面 · safety论文趋势 2023–2026

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。

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 42 篇，第 1 / 1 页

1. ICLR 2026A2D: Any-Order, Any-Step Safety Alignment for Diffusion Language Models
2. ICLR 2026Any-Depth Alignment: Unlocking Innate Safety Alignment of LLMs to Any-Depth
3. ICLR 2026Beyond Linear Probes: Dynamic Safety Monitoring for Language Models
4. ICLR 2026Bi-Criteria Metric Distortion
5. ICLR 2026Erase or Hide? Suppressing Spurious Unlearning Neurons for Robust Unlearning
6. ICLR 2026GAVEL: Towards Rule-Based Safety through Activation Monitoring
7. ICLR 2026GradShield: Alignment Preserving Finetuning
8. ICLR 2026GuardAlign: Robust Safety Alignment in Multimodal Large Language Models
9. ICLR 2026Invisible Safety Threat: Malicious Finetuning for LLM via Steganography
10. ICLR 2026OffTopicEval: When Large Language Models Enter the Wrong Chat, Almost Always!
11. ICLR 2026On the Impossibility of Separating Intelligence from Judgment: The Computational Intractability of Filtering for AI Alignment
12. ICLR 2026PRISON: Unmasking the Criminal Potential of Large Language Models
13. ICLR 2026ProSafePrune: Projected Safety Pruning for Mitigating Over-Refusal in LLMs
14. ICLR 2026Reasoned Safety Alignment: Ensuring Jailbreak Defense via Answer-Then-Check
15. ICLR 2026Reward Models Inherit Value Biases from Pretraining
16. ICLR 2026Safety Subspaces are Not Linearly Distinct: A Fine-Tuning Case Study
17. ICLR 2026Social Agents: Collective Intelligence Improves LLM Predictions
18. ICLR 2026Steering Language Models with Weight Arithmetic
19. ICLR 2026Superficial Safety Alignment Hypothesis
20. ICLR 2026When Style Breaks Safety: Defending LLMs Against Superficial Style Alignment
21. ICLR 2025A Black Swan Hypothesis: The Role of Human Irrationality in AI Safety
22. ICLR 2025Breach By A Thousand Leaks: Unsafe Information Leakage in 'Safe' AI Responses
23. ICLR 2025Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirements
24. ICLR 2025Does Safety Training of LLMs Generalize to Semantically Related Natural Prompts?
25. ICLR 2025Durable Quantization Conditioned Misalignment Attack on Large Language Models
26. ICLR 2025Jailbreak Antidote: Runtime Safety-Utility Balance via Sparse Representation Adjustment in Large Language Models
27. ICLR 2025On the Role of Attention Heads in Large Language Model Safety
28. ICLR 2025Programming Refusal with Conditional Activation Steering
29. ICLR 2025Safety Alignment Should be Made More Than Just a Few Tokens Deep
30. ICLR 2025Safety Layers in Aligned Large Language Models: The Key to LLM Security
31. ICLR 2025The Hidden Cost of Waiting for Accurate Predictions
32. ICLR 2024Fine-tuning Aligned Language Models Compromises Safety, Even When Users Do Not Intend To!
33. ICLR 2024Gaining Wisdom from Setbacks: Aligning Large Language Models via Mistake Analysis
34. ICLR 2024GPT-4 Is Too Smart To Be Safe: Stealthy Chat with LLMs via Cipher
35. ICLR 2024Multilingual Jailbreak Challenges in Large Language Models
36. ICLR 2024Safety-Tuned LLaMAs: Lessons From Improving the Safety of Large Language Models that Follow Instructions
37. ICLR 2024Social-Transmotion: Promptable Human Trajectory Prediction
38. ICLR 2024Some Fundamental Aspects about Lipschitz Continuity of Neural Networks
39. ICLR 2024The Alignment Problem from a Deep Learning Perspective
40. ICLR 2024The Devil is in the Neurons: Interpreting and Mitigating Social Biases in Language Models
41. ICLR 2024Training Socially Aligned Language Models on Simulated Social Interactions
42. ICLR 2024Unprocessing Seven Years of Algorithmic Fairness

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。