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 688 篇，第 2 / 14 页

51. ICML 2026Calibrating Conservatism for Scalable Oversight
52. ICML 2026CapBencher: Give Your LLM Benchmark a Built-in Alarm for Test-Set Overfitting
53. ICML 2026Certified Circuits: Stability Guarantees for Mechanistic Circuits
54. ICML 2026Collaborative Disagreement Resolution for Scalable Oversight
55. ICML 2026Collaborative Threshold Watermarking
56. ICML 2026Concept Concentration for Faithful Representation Intervention
57. ICML 2026Consistency Training Can Entrench Misalignment
58. ICML 2026Constitutional Black-Box Monitoring for Scheming in LLM Agents
59. ICML 2026Contrastive Reasoning Alignment: Reinforcement Learning from Hidden Representations
60. ICML 2026CoopEval: Benchmarking Cooperation-Sustaining Mechanisms and LLM Agents in Social Dilemmas
61. ICML 2026Copyright-Bench: Agentic Evaluation of Copyright Law Compliance
62. ICML 2026CORE: Conflict-Oriented Reasoning for General Multimodal Manipulation Detection
63. ICML 2026Corrigibility Transformation: Constructing Goals That Accept Updates
64. ICML 2026CoT is Not the Chain of Truth: An Empirical Internal Analysis of Reasoning LLMs for Fake News Generation
65. ICML 2026D-Judge: Disrupting Multi-Turn Jailbreaks using Semantics-Preserving Output Rewriting
66. ICML 2026DDIM Inversion as a Perturbation Amplifier: Breaking Mimicry Protection via Reconstruction Error Minimization
67. ICML 2026Debate with Images: Detecting Deceptive Behaviors in Multimodal Large Language Models
68. ICML 2026Decomposing Query-Key Feature Interactions Using Contrastive Covariances
69. ICML 2026Detecting Fluent Optimization-Based Adversarial Prompts via Sequential Entropy Changes
70. ICML 2026Diagnosing the Reliability of LLM-as-a-Judge via Item Response Theory
71. ICML 2026Dimensional Collapse in Transformer Attention Outputs: A Challenge for Sparse Dictionary Learning
72. ICML 2026Discovering Implicit Large Language Model Alignment Objectives
73. ICML 2026Dismantling Pathological Shortcuts: A Causal Framework for Faithful LVLM Decoding
74. ICML 2026Dissect and Prune: Enhancing Robustness in AI-Generated Image Detection
75. ICML 2026Dissecting the Safety Circuit: Neuronal Intervention for Transferable Adversarial Attacks on VLMs
76. ICML 2026Divergence Decoding: Inference-Time Unlearning via Auxiliary Models
77. ICML 2026Do Language Models Track Entities Across State Changes?
78. ICML 2026Don't Walk the Line: Boundary Guidance for Filtered Generation
79. ICML 2026Dual-branch Robust Unlearnable Examples
80. ICML 2026Efficient LLM Moderation with Multi-Layer Latent Prototypes
81. ICML 2026EMBGuard: Constructing Hazard-Aware Guardrails for Safe Planning in Embodied Agents
82. ICML 2026Emergence of Biased Consensus in Multi-Agent LLM Debates
83. ICML 2026Emergence of Hierarchical Emotion Organization in Large Language Models
84. ICML 2026Emergent Communication Under Misinformation
85. ICML 2026Enhancing Conformal Prediction via Class Similarity
86. ICML 2026Ensembling Sparse Autoencoders
87. ICML 2026Estimating Tail Risks in Language Model Output Distributions
88. ICML 2026Exactly Computing do-Shapley Values
89. ICML 2026ExPLAIND: Unifying Model, Data, and Training Attribution to Study Model Behavior
90. ICML 2026Exploration Hacking: Can LLMs Learn to Resist RL Training?
91. ICML 2026Exploring Nonlinear Pathway in Parameter Space for Machine Unlearning
92. ICML 2026Eyes-on-Me: Scalable RAG Poisoning through Transferable Attention-Steering Attractors
93. ICML 2026Fair Decisions from Calibrated Scores: Achieving Optimal Classification While Satisfying Sufficiency
94. ICML 2026Faithful Mobile GUI Agents with Guided Advantage Estimator
95. ICML 2026Falling Trees: A Model Class for Interpretable Risk Prioritization
96. ICML 2026Finding Most Influential Sets
97. ICML 2026FLARE-AI: Flaw Reporting for AI
98. ICML 2026Fleet: Few Shots Lead Effective AI-generated Image Detection
99. ICML 2026FoeGlass: Simple In-Context Learning Is Enough for Red Teaming Audio Deepfake Detectors
100. ICML 2026ForceForget: Reinforcement Concept Removal for Enhancing Safety in Text-to-Image Models

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。