可微推理论文趋势 2023–2026

主要结论

• 可微推理论文从 2023 年的 31 篇增长到 2026 年的 79 篇，三年间扩大到约 2.5 倍。
• NeurIPS 是发可微推理论文最多的会议，2023–2026 年共 65 篇，占四个会议总量的 35%。
• 交叉方向里涨得最快的是数学推理：同时属于两个主题的论文从 2023 年的 3 篇增至 2026 年的 29 篇。
• 在「神经符号」的 5 个主题里，它的 2026 年论文数排第 3；同组增长最快的数学推理是 10 倍，可微推理是 2.5 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 2026Beyond MSE: Differentiable Complexity Priors for Structure-Preserving Neural Denoising
• ICML 2026Differentiable Conformal Training for LLM Reasoning Factuality
• ICLR 2026$\boldsymbol{\partial^\infty}$-Grid: A Neural Differential Equation Solver with Differentiable Feature Grids
• CVPR 2026D-Convexity A Unified Differentiable Convex Shape Prior via Quasi-Concavity for
• NeurIPS 2025Consistency Conditions for Differentiable Surrogate Losses
• ICML 2025DiLQR: Differentiable Iterative Linear Quadratic Regulator via Implicit Differentiation
• ICLR 2025Analytic DAG Constraints for Differentiable DAG Learning
• CVPR 2025Differentiable Inverse Rendering with Interpretable Basis BRDFs
• NeurIPS 2024DeiSAM: Segment Anything with Deictic Prompting
• ICML 2024A Differentiable Partially Observable Generalized Linear Model with Forward-Backward Message Passing
• ICLR 2024Generalized Neural Sorting Networks with Error-Free Differentiable Swap Functions
• CVPR 2024Differentiable Display Photometric Stereo
• NeurIPS 2023Differentiable Clustering with Perturbed Spanning Forests
• ICML 2023End-to-end Differentiable Clustering with Associative Memories
• ICLR 2023Learning Group Importance using the Differentiable Hypergeometric Distribution
• CVPR 2023Deep Learning of Partial Graph Matching via Differentiable Top-K
• NeurIPS 2026Differentiable Exact Learning of Algorithms
• ICML 2026Hide&Seek: Learning to Explain in an End-to-End Differentiable Network
• ICLR 2026Tversky Neural Networks: Psychologically Plausible Deep Learning with Differentiable Tversky Similarity
• CVPR 2026DiffBMP Differentiable Rendering with Bitmap Primitives

全部可微推理论文

共 78 篇，第 1 / 2 页

1. CVPR 2026D-Convexity A Unified Differentiable Convex Shape Prior via Quasi-Concavity for
2. CVPR 2026DiffBMP Differentiable Rendering with Bitmap Primitives
3. CVPR 2026Differentiable Adaptive 4D Structured Illumination for Joint Capture of Shape
4. CVPR 2026Differentiable Laplacian Matrix Guided Superpixel Segmentation
5. CVPR 2026DiffSoup Direct Differentiable Rasterization of Triangle Soup for Extreme Radiance
6. CVPR 2026GeoCoT Towards Reliable Remote Sensing Reasoning with Manifold Perspective
7. CVPR 2026Geoint-R1 Formalizing Multimodal Geometric Reasoning with Dynamic Auxiliary Constructions
8. CVPR 2026Hermite Radial Basis Function for Surface Reconstruction via Differentiable Rendering
9. CVPR 2026Learning Differentiable Hierarchies in 3D Gaussian Splatting
10. CVPR 2026Lens Component Deletion based on Differentiable Ray Tracing
11. CVPR 2026MeshSplatting Differentiable Rendering with Opaque Meshes
12. CVPR 2026MVInverse Feed-forward Multiview Inverse Rendering in Seconds
13. CVPR 2026Parallelised Differentiable Straightest Geodesics for 3D Meshes
14. CVPR 2026Perceiving the Near Reasoning the Distant Coherent Long-Horizon Trajectory Prediction
15. CVPR 2026Proof-of-Perception Certified Tool-Using Multimodal Reasoning with Compositional Conformal Guarantees
16. CVPR 2026SPREAD Spatial-Physical REasoning via geometry Aware Diffusion
17. CVPR 2026Think with 3D Geometric Imagination Grounded Spatial Reasoning from Limited
18. CVPR 2026Towards Reasoning-Preserving Unlearning in Multimodal Large Language Models
19. CVPR 2026Visual-RRT Finding Paths toward Visual-Goals via Differentiable Rendering
20. CVPR 2026When Visualizing is the First Step to Reasoning MIRA a
21. ICLR 2026$\boldsymbol{\partial^\infty}$-Grid: A Neural Differential Equation Solver with Differentiable Feature Grids
22. ICLR 2026Accelerated Learning with Linear Temporal Logic using Differentiable Simulation
23. ICLR 2026ActivationReasoning: Logical Reasoning in Latent Activation Spaces
24. ICLR 2026CyclicReflex: Improving Reasoning Models via Cyclical Reflection Token Scheduling
25. ICLR 2026DiVeQ: Differentiable Vector Quantization Using the Reparameterization Trick
26. ICLR 2026Efficient Reasoning with Balanced Thinking
27. ICLR 2026Efficient Spatially-Variant Convolution via Differentiable Sparse Kernel Complex
28. ICLR 2026KaVa: Latent Reasoning via Compressed KV-Cache Distillation
29. ICLR 2026Learning to Reason via Mixture-of-Thought for Logical Reasoning
30. ICLR 2026Light Differentiable Logic Gate Networks
31. ICLR 2026NePTune: A Neuro-Pythonic Framework for Tunable Compositional Reasoning on Vision-Language
32. ICLR 2026Probability Distributions Computed by Autoregressive Transformers
33. ICLR 2026RAIN-Merging: A Gradient-Free Method to Enhance Instruction Following in Large Reasoning Models with Preserved Thinking Format
34. ICLR 2026RuleReasoner: Reinforced Rule-based Reasoning via Domain-aware Dynamic Sampling
35. ICLR 2026The Geometry of Reasoning: Flowing Logics in Representation Space
36. ICLR 2026Tversky Neural Networks: Psychologically Plausible Deep Learning with Differentiable Tversky Similarity
37. ICML 2026Abductive Reasoning with Probabilistic Commonsense
38. ICML 2026AdaSplash-2: Faster Differentiable Sparse Attention
39. ICML 2026Align Forward, Adapt Backward: Closing the Discretization Gap in Logic Gate Networks
40. ICML 2026d$^2$p: Structured Soft Attention Is All You Need
41. ICML 2026Deterministic Differentiable Structured Pruning for Large Language Models
42. ICML 2026Differentiable Conformal Training for LLM Reasoning Factuality
43. ICML 2026Differentiable Weightless Controllers: Learning Logic Circuits for Continuous Control
44. ICML 2026Diffusion Differentiable Resampling
45. ICML 2026Equilibrium Reasoners: Learning Attractors Enables Scalable Reasoning
46. ICML 2026Hide&Seek: Learning to Explain in an End-to-End Differentiable Network
47. ICML 2026SoftJAX & SoftTorch: Empowering Automatic Differentiation Libraries with Informative Gradients
48. ICML 2026TT-Sparse: Learning Sparse Rule Models with Differentiable Truth Tables
49. ICML 2026TVDRNet: Text-driven Viewpoint Optimization via Differentiable Rendering for 3D Reasoning Segmentation
50. ICML 2026Two-Stage Unit Tying for Simplifying Differentiable Logic Gate Networks

常见问题

2023–2026 年四大会议一共有多少篇可微推理论文？

共 185 篇，按年份分别是 2023 年 31 篇、2024 年 29 篇、2025 年 46 篇、2026 年 79 篇。

NeurIPS 2026 有多少篇可微推理论文？

NeurIPS 2026 有 28 篇可微推理论文，上一届是 22 篇。

ICLR 2026 有多少篇可微推理论文？

ICLR 2026 有 16 篇可微推理论文，上一届是 13 篇。

ICML 2026 有多少篇可微推理论文？

ICML 2026 有 14 篇可微推理论文，上一届是 5 篇。

CVPR 2026 有多少篇可微推理论文？

CVPR 2026 有 21 篇可微推理论文，上一届是 6 篇。

哪个会议的可微推理论文最多？

NeurIPS，2023–2026 年共 65 篇。