因果推荐论文趋势 2023–2026

主要结论

• 因果推荐论文从 2023 年的 83 篇增长到 2026 年的 190 篇，三年间扩大到约 2.3 倍。
• NeurIPS 是发因果推荐论文最多的会议，2023–2026 年共 227 篇，占四个会议总量的 41%。
• 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 21 篇。
• 在「推荐与广告」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的点击率预测是 2.6 倍，因果推荐是 2.3 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 2026A Characterization of Latent Variable Causal Models Consistent with Observational Data
• ICML 2026A Causal Decomposition Approach for Fair Contextual Multi-Armed Bandits
• ICLR 2026ActiveCQ: Active Estimation of Causal Quantities
• CVPR 2026Breaking Spurious Correlations Uncertainty-Driven Causal Transformers for AU Detection
• NeurIPS 2025CCL: Causal-aware In-context Learning for Out-of-Distribution Generalization
• ICML 2025A Geometric Approach to Personalized Recommendation with Set-Theoretic Constraints Using Box Embeddings
• ICLR 2025Causal Concept Graph Models: Beyond Causal Opacity in Deep Learning
• CVPR 2025Image Quality Assessment: Investigating Causal Perceptual Effects with Abductive Counterfactual Inference
• NeurIPS 2024Causal Bandits for Linear Structural Equation Models
• ICML 2024A Sparsity Principle for Partially Observable Causal Representation Learning
• ICLR 2024Causal Inference with Conditional Front-Door Adjustment and Identifiable Variational Autoencoder
• CVPR 2024MuseChat: A Conversational Music Recommendation System for Videos
• NeurIPS 2023A Causal Framework for Decomposing Spurious Variations
• ICML 2023B-Learner: Quasi-Oracle Bounds on Heterogeneous Causal Effects Under Hidden Confounding
• ICLR 2023Calibration Matters: Tackling Maximization Bias in Large-scale Advertising Recommendation Systems
• CVPR 2023CafeBoost: Causal Feature Boost To Eliminate Task-Induced Bias for Class Incremental Learning
• NeurIPS 2026Bridging Image Restoration and Recognition via Causal Mediated Unrolling
• ICML 2026Causal Disentangled Anchor Learning for Scalable Fair Multi-view Clustering
• ICLR 2026Addressing divergent representations from causal interventions on neural networks
• CVPR 2026Bridge Basis-Driven Causal Inference Marries VFMs for Domain Generalization

全部因果推荐论文

共 548 篇，第 2 / 11 页

51. ICLR 2026Query-Specific Causal Graph Pruning Under Tiered Knowledge
52. ICLR 2026Score-based Greedy Search for Structure Identification of Partially Observed Linear Causal Models
53. ICLR 2026Steering Diffusion Models Towards Credible Content Recommendation
54. ICLR 2026Stochastic Neural Networks for Causal Inference with Missing Confounders
55. ICLR 2026SurvHTE-Bench: A Benchmark for Heterogeneous Treatment Effect Estimation in Survival Analysis
56. ICLR 2026Testing Most Influential Sets
57. ICLR 2026Token-Efficient Long-Term Interest Sketching and Internalized Reasoning for LLM-based Recommendation
58. ICLR 2026Topological Causal Effects
59. ICLR 2026Truthful or Fabricated? Using Causal Attribution to Mitigate Reward Hacking in Explanations
60. ICML 2026A Causal Decomposition Approach for Fair Contextual Multi-Armed Bandits
61. ICML 2026A Recursive Decomposition Framework for Causal Structure Learning in the Presence of Latent Variables
62. ICML 2026Adaptive Estimation and Inference in Semi-parametric Heterogeneous Clustered Multitask Learning via Neyman Orthogonality
63. ICML 2026AgentSelect: Benchmark for Narrative Query-to-Agent Recommendation
64. ICML 2026Anti-causal domain generalization: Leveraging unlabeled data
65. ICML 2026Any-Order GPT as Masked Diffusion Model: Decoupling Formulation and Architecture
66. ICML 2026Caracal: Causal Architecture via Spectral Mixing
67. ICML 2026Causal Attention with Lookahead Keys
68. ICML 2026Causal Direct Preference Optimization for Distributionally Robust Generative Recommendation
69. ICML 2026Causal Disentangled Anchor Learning for Scalable Fair Multi-view Clustering
70. ICML 2026Causal Effect Identifiability in the Presence of Latent Confounders Without Auxiliary Variables
71. ICML 2026Causal Feature Learning via Generalized Rayleigh Quotients
72. ICML 2026Causal Fine-Tuning under Latent Confounded Shift
73. ICML 2026Causal Flow Q-Learning for Robust Offline Reinforcement Learning
74. ICML 2026Causal Identification from Counterfactual Data: Completeness and Bounding Results
75. ICML 2026Causal Matrix Completion under Multiple Treatments via Mixed Synthetic Nearest Neighbors
76. ICML 2026Causal Preference Elicitation
77. ICML 2026Causal Representation Learning with Optimal Compression and Complex Treatments
78. ICML 2026Causal-EPIG: Causally Aligned Active CATE Estimation
79. ICML 2026Conditional Distributional Treatment Effects: Doubly Robust Estimation and Testing
80. ICML 2026CORE-MTL: Rethinking Gradient Balancing via Causal Orthogonal Representations
81. ICML 2026Dissecting Causal Mechanism Shifts via FANS: Function And Noise Separation
82. ICML 2026Distributionally Robust Causal Abstractions
83. ICML 2026Do-Prompt: Causal Interventions Meet Variational Prompt Bottlenecks
84. ICML 2026Evaluating Bivariate Causal Statements Based on Mutual Compatibility
85. ICML 2026Factored Classifier-Free Guidance
86. ICML 2026Fast and Accurate Causal Parallel Decoding using Jacobi Forcing
87. ICML 2026Federated Causal Inference on Multi-Site Observational Data via Propensity Score Aggregation
88. ICML 2026Formalizing and Falsifying Causal Pathways of Rare Events
89. ICML 2026Frequentist Consistency of Prior-Data Fitted Networks for Causal Inference
90. ICML 2026Function-Valued Causal Influence in Nonlinear Time Series
91. ICML 2026Gateways to Tractability for Satisfiability in Pearl’s Causal Hierarchy
92. ICML 2026Gradient-Based Causal Tree Ensembles: A Backbone Architecture for Heterogeneous Treatment Effects
93. ICML 2026Hyperbolic RQ-VAE enhanced Generative Recommendation with Differential-Length Codebook Strategy
94. ICML 2026Identifying Partially Observed Causal Models from Heterogeneous/Nonstationary Data
95. ICML 2026Interventional Processes For Causal Uncertainty Quantification
96. ICML 2026Linear Causal Representation Learning by Topological Ordering, Pruning, and Disentanglement
97. ICML 2026Many Experiments, Few Repetitions, Unpaired Data, and Sparse Effects: Is Causal Inference Possible?
98. ICML 2026Negative Sampling From the Ground Up: A Redesign for Recommendation
99. ICML 2026On the Identifiability of Poisson Branching Structural Causal Model Under Latent Confounding
100. ICML 2026Relational Structural Causal Models

常见问题

2023–2026 年四大会议一共有多少篇因果推荐论文？

共 548 篇，按年份分别是 2023 年 83 篇、2024 年 123 篇、2025 年 152 篇、2026 年 190 篇。

NeurIPS 2026 有多少篇因果推荐论文？

NeurIPS 2026 有 75 篇因果推荐论文，上一届是 69 篇。

ICLR 2026 有多少篇因果推荐论文？

ICLR 2026 有 43 篇因果推荐论文，上一届是 32 篇。

ICML 2026 有多少篇因果推荐论文？

ICML 2026 有 56 篇因果推荐论文，上一届是 49 篇。

CVPR 2026 有多少篇因果推荐论文？

CVPR 2026 有 16 篇因果推荐论文，上一届是 2 篇。

哪个会议的因果推荐论文最多？

NeurIPS，2023–2026 年共 227 篇。