Deep learning · causal inference papers, 2023–2026

Key findings

• Deep learning · causal inference fell from 108 papers in 2023 to 66 in 2026, so interest is cooling.
• NeurIPS publishes the most Deep learning · causal inference work: 196 papers from 2023 to 2026, 62% of the four-venue total.
• The fastest-growing overlap is with Causal Recommendation: papers tagged with both rose from 27 in 2023 to 55 in 2026.
• Among the 11 topics under Deep Learning, it ranks #10 by 2026 paper count; the fastest grower in the group, Diffusion Models, is up 4.1x versus 0.6x here.

Representative papers

Spread across years and venues so every edition shows up

• NeurIPS 2026From Groups to Rings: Causal Evidence for Algebraic Decomposition in Grokked Transformers
• ICML 2026Towards a Holistic Understanding of Selection Bias for Causal Effect Identification
• ICLR 2026Causal-Steer: Disentangled Continuous Style Control without Parallel Corpora
• CVPR 2026Breaking Spurious Correlations Uncertainty-Driven Causal Transformers for AU Detection
• NeurIPS 2025CCL: Causal-aware In-context Learning for Out-of-Distribution Generalization
• ICML 2025Position: Probabilistic Modelling is Sufficient for Causal Inference
• ICLR 2025Robust Root Cause Diagnosis using In-Distribution Interventions
• CVPR 2025Image Quality Assessment: Investigating Causal Perceptual Effects with Abductive Counterfactual Inference
• NeurIPS 2024A Non-parametric Direct Learning Approach to Heterogeneous Treatment Effect Estimation under Unmeasured Confounding
• ICML 2024Collaborative Heterogeneous Causal Inference Beyond Meta-analysis
• ICLR 2024Causal-StoNet: Causal Inference for High-Dimensional Complex Data
• CVPR 2024Bi-Causal: Group Activity Recognition via Bidirectional Causality
• NeurIPS 2023A Cross-Moment Approach for Causal Effect Estimation
• ICML 2023Proximal Causal Learning of Conditional Average Treatment Effects
• ICLR 2023GEASS: Neural causal feature selection for high-dimensional biological data
• CVPR 2023CafeBoost: Causal Feature Boost To Eliminate Task-Induced Bias for Class Incremental Learning
• NeurIPS 2026A Characterization of Latent Variable Causal Models Consistent with Observational Data
• ICML 2026Caracal: Causal Architecture via Spectral Mixing
• ICLR 2026Counterfactual Structural Causal Bandits
• CVPR 2026Bridge Basis-Driven Causal Inference Marries VFMs for Domain Generalization

All Deep learning · causal inference papers

316 papers, page 1 of 7

1. CVPR 2026Breaking Spurious Correlations Uncertainty-Driven Causal Transformers for AU Detection
2. CVPR 2026Bridge Basis-Driven Causal Inference Marries VFMs for Domain Generalization
3. CVPR 2026GeneVAR Causal MeanFlow for Autoregressive Gene-to-WSI Tile Synthesis
4. CVPR 2026Learning by Analogy A Causal Framework for Compositional Generalization
5. ICLR 2026Addressing divergent representations from causal interventions on neural networks
6. ICLR 2026Causal-Steer: Disentangled Continuous Style Control without Parallel Corpora
7. ICLR 2026Coarse-to-Fine Learning of Dynamic Causal Structures
8. ICLR 2026Contextual Causal Bayesian Optimisation
9. ICLR 2026Counterfactual Structural Causal Bandits
10. ICLR 2026Exploratory Causal Inference in SAEnce
11. ICLR 2026On Measuring Influence in Avoiding Undesired Future
12. ICLR 2026Topological Causal Effects
13. ICML 2026ARC-Decode: Accelerated Decoding with Risk-Bounded Acceptance
14. ICML 2026Beyond Reactivity: Proactive Adaptive Conformal Inference for Online LLM Factuality
15. ICML 2026Caracal: Causal Architecture via Spectral Mixing
16. ICML 2026Causal Attention with Lookahead Keys
17. ICML 2026Causal Effect Identifiability in the Presence of Latent Confounders Without Auxiliary Variables
18. ICML 2026Causal Feature Learning via Generalized Rayleigh Quotients
19. ICML 2026Causal Matrix Completion under Multiple Treatments via Mixed Synthetic Nearest Neighbors
20. ICML 2026Causal Preference Elicitation
21. ICML 2026Cerebellar-Inspired Residual Control for Fault Recovery: From Inference-Time Adaptation to Structural Consolidation
22. ICML 2026Correctness-Optimized Residual Activation Lens (CORAL): Transferrable and Calibration-Aware Inference-Time Steering
23. ICML 2026Distributionally Robust Causal Abstractions
24. ICML 2026Evaluating Bivariate Causal Statements Based on Mutual Compatibility
25. ICML 2026Federated Causal Inference on Multi-Site Observational Data via Propensity Score Aggregation
26. ICML 2026Frequentist Consistency of Prior-Data Fitted Networks for Causal Inference
27. ICML 2026Identifying Partially Observed Causal Models from Heterogeneous/Nonstationary Data
28. ICML 2026Reward Shaping for (Inference-Time) Alignment: A Stackelberg Game Perspective
29. ICML 2026Safe Reinforcement Learning with Preference-based Constraint Inference
30. ICML 2026SCOUT: Cyclic Causal Discovery Under Soft Interventions with Unknown Targets
31. ICML 2026Sinkhorn Treatment Effects: A Causal Optimal Transport Measure
32. ICML 2026SparseInfer: Accelerating Large Language Model Inference with Semantics-Inspired Adaptive Sparse Activation
33. ICML 2026Stem: Rethinking Causal Information Flow in Sparse Attention
34. ICML 2026Tailoring Strictly Proper Scoring Rules for Downstream Tasks: An Application to Causal Inference
35. ICML 2026Target-Driven Policy Optimization for Sequential Counterfactual Outcome Control
36. ICML 2026TimeSAE: Causal Sparse Decoding for Faithful Explanations of Black-Box Time Series Models
37. ICML 2026Towards a Holistic Understanding of Selection Bias for Causal Effect Identification
38. ICML 2026Uncovering Bias Mechanisms in Observational Studies
39. NeurIPS 2026A Characterization of Latent Variable Causal Models Consistent with Observational Data
40. NeurIPS 2026A Faster Algorithm for the Half-Trek Criterion in Structural Causal Models
41. NeurIPS 2026CAST: Causal Anchored Simplex Transport for Distribution-Valued Time Series
42. NeurIPS 2026Causal Abstractions, Categorically Unified
43. NeurIPS 2026Causal Effect Identification with a Single Agnostic Proxy
44. NeurIPS 2026Causal Gradient Steering: Exposing Shortcut Gradients by Destroying Causal Signal
45. NeurIPS 2026Causal learning with the invariance principle
46. NeurIPS 2026Causal Survival Forests with Negative Controls
47. NeurIPS 2026Characterizing Underrepresentation in Generalizing Causal Survival Estimates
48. NeurIPS 2026Confounding-Aware Client Selection in Federated Learning via Causal Mediation Analysis
49. NeurIPS 2026Decision Path Tracing for Causal Analysis in Transformers
50. NeurIPS 2026Decomposing Effects in Neural Causal Models

FAQ

How many Deep learning · causal inference papers were published at NeurIPS, ICLR, ICML and CVPR from 2023 to 2026?

316 in total: 108 in 2023, 108 in 2024, 34 in 2025, 66 in 2026.

How many Deep learning · causal inference papers are at NeurIPS 2026?

NeurIPS 2026 has 28 Deep learning · causal inference papers, versus 9 the year before.

How many Deep learning · causal inference papers are at ICLR 2026?

ICLR 2026 has 8 Deep learning · causal inference papers, versus 6 the year before.

How many Deep learning · causal inference papers are at ICML 2026?

ICML 2026 has 26 Deep learning · causal inference papers, versus 18 the year before.

How many Deep learning · causal inference papers are at CVPR 2026?

CVPR 2026 has 4 Deep learning · causal inference papers, versus 1 the year before.

Which conference publishes the most Deep learning · causal inference papers?

NeurIPS, with 196 papers from 2023 to 2026.