| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Multivariate distribution estimation | Synthetic Data I | ED0.023 | 35 | |
| Tool Use | Synthetic Data (test) | Task Accuracy92.29 | 24 | |
| Causal Discovery | Synthetic Data | Runtime111 | 21 | |
| Symbolic Reconstruction | Synthetic Data Interact Mode (test) | Reconstruction Accuracy108.04 | 20 | |
| Canonical Correlation Analysis | Synthetic Data | Aggregate Disparity (Delta_sum,r)2.2304 | 20 | |
| 3D Shape Completion | Synthetic data (test) | Chamfer Distance (CD)0.61 | 19 | |
| Conformal Prediction | Synthetic Data sample size 5000 (test) | Marginal Coverage90.7 | 16 | |
| Inference Efficiency Analysis | Synthetic data n=100 fields, k=8 embedding size (evaluation) | Latency (ms)0.0012 | 13 | |
| Bandit Learning | Synthetic Data Attack | Regret (Mean)1,382.96 | 10 | |
| Sparsity Recovery | Synthetic data (SNR=20, 0.2 correlation, 0.05 true density) v1 (train) | TDR100 | 9 | |
| Multi-instance 3D registration | Synthetic data | MHR53.39 | 8 | |
| Bandit Learning | Synthetic Data Benign | Mean Regret318.52 | 8 | |
| Hyperspectral Unmixing | Synthetic Data | aRMSE2.15 | 8 | |
| Multi-view fluid reconstruction | Synthetic data (test) | rhoH RMSE1.3 | 8 | |
| Causal Effect Estimation | Synthetic data | PEHE (gamma=0)0.43 | 7 | |
| Causal Discovery | Synthetic Data |V|=5 | F1 Score81 | 6 | |
| Causal Discovery | Synthetic Data (|V|=15) | F1 Score68 | 6 | |
| Causal Discovery | Synthetic Data |V|=10 | F1 Score66 | 6 | |
| Classification | Synthetic data (SNR=20, 0.2 correlation, 0.05 true density) v1 (test) | Accuracy0.9 | 6 | |
| Spike-based Image Reconstruction | Synthetic Data Light intensity (64) | PSNR24.46 | 6 | |
| Spike-based Image Reconstruction | Synthetic Data Light intensity (32) | PSNR23.66 | 6 | |
| Spike-based Image Reconstruction | Synthetic Data Light intensity (16) | PSNR19.38 | 6 | |
| Text Generation | Synthetic Data Length 40 | Oracle NLL6.14 | 6 | |
| Text Generation | Synthetic Data Length 20 | Oracle NLL5.67 | 6 | |
| Symbolic Regression | Synthetic Data Hybrid Mode | Mean MSE (d=10)0.0423 | 5 |