Research map / Label-efficient and robust learning / Test-time adaptation
Multi-Task Test-time Adaptation via Gradient Consensus and Plasticity Constraint
Zhong Ye, Yu Hu, Zhenguo Yang
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 |
|---|---|---|---|
| add gradient consensus plus plasticity constraint to M-TENT (ablation) | M-TENT · PlantData (avg of 6 shifts) | Mean I&O F1: 90.52 -> 91.57 | Improved |
| add gradient consensus plus plasticity constraint to M-TENT (ablation) | M-TENT · CelebA (avg of 6 shifts) | Mean I&O F1: 93.35 -> 93.37 | No clear effect |
| add only one of the two modules (ablation) | M-TENT · PlantData (avg of 6 shifts) | Mean I&O F1: 90.52 -> 90.66 / 90.57 | No clear effect |
| add both modules, average across datasets (ablation) | M-TENT · PlantData and CelebA | Average Mean I&O across datasets: 91.94 -> 92.47 | Improved |
| use CoCo-MT-TTA instead of M-TENT | multi-task TTA · PlantData Gaussian noise | Mean I&O F1: 92.17 -> 93.12 | Improved |
| use CoCo-MT-TTA instead of EATA | multi-task TTA · PlantData Impulse noise | Mean I&O F1: 88.99 -> 87.25 | Hurt |
| use CoCo-MT-TTA instead of EATA | multi-task TTA · PlantData Contrast | Mean I&O F1: 77.85 -> 77.52 | No clear effect |
| apply test-time adaptation (M-TENT) without ID constraint | multi-task model · PlantData | F1 on ID samples | Hurt |
| use ActMAD adaptation | multi-task model · PlantData | Mean I&O F1: 75.55 -> 34.44 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ActMAD — ActMAD: Activation Matching To Align Distributions for Test-Time-Training (CVPR 2023) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Towards Stable Test-time Adaptation in Dynamic Wild World (ICLR 2023 · similarity 0.89)
- Multi-Hypothesis Test-Time Adaptation to Mitigate Underspecification (ECCV 2026 · similarity 0.89)
- A Layer Selection Approach to Test Time Adaptation (AAAI 2025 · similarity 0.89)
- Bayesian Weight Enhancement with Steady-State Adaptation for Test-time Adaptation in Dynamic Environments (ICML 2025 · similarity 0.87)
- Towards Real-World Test-Time Adaptation: Tri-net Self-Training with Balanced Normalization (AAAI 2024 · similarity 0.87)
- Test-Time Adaptation with Binary Feedback (ICML 2025 · similarity 0.87)
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.
