Research map / Label-efficient and robust learning / Test-time adaptation
Predictor Combination at Test Time
Kwang In Kim, James Tompkin, Christian Richardt
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use GP predictive variances instead of unit covariance for reference predictors (ablation) | Predictor combination · MOCAP (8 targets) | average error rate: reduced the average error rate by 11.34% | Improved |
| combine reference predictors by manifold denoising on CAESAR | Ind GP predictor · CAESAR (arm length, foot length) | mean squared error | Mixed |
| combine reference predictors on SARCOS | Ind GP predictor · SARCOS | mean squared error | Mixed |
| combine reference predictors on MOCAP | Ind GP predictor · MOCAP (8 targets) | mean squared error | Improved |
| apply graph-Laplacian multi-task learning to facial landmark SVM detectors | SVM landmark detectors · Landmarks (BioID) | landmark error in pixels | Hurt |
| combine predictors from 138 other schools with proposed method | Ind GP predictor · School (20 labeled points) | mean squared error: 11.86 -> 11.24 | Improved |
| use uniform graph Laplacian GL1 instead of proposed method | Predictor combination · School (20 labeled points) | mean squared error: 11.24 -> 10.80 | Improved |
Closest papers by meaning
- Joint Manifold Diffusion for Combining Predictions on Decoupled Observations (CVPR 2019 · similarity 0.90)
- Fast, Accurate Manifold Denoising by Tunneling Riemannian Optimization (ICML 2025 · similarity 0.82)
- Nonparametric In-Context Learning under Growing Geometric Complexity: Minimax Optimality and Local Geometry-Adaptivity of Transformers (NeurIPS 2026 · similarity 0.82)
- Task-Induced Riemannian Metrics for Vision Transformer Feature Spaces (NeurIPS 2026 · similarity 0.82)
- Analysing Multi-Task Regression via Random Matrix Theory with Application to Time Series Forecasting (NeurIPS 2024 · similarity 0.82)
- Mixed-Sample SGD: an End-to-end Analysis of Supervised Transfer Learning (NeurIPS 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
