Research map / Label-efficient and robust learning / Test-time adaptation
Rethinking Entropy in Test-Time Adaptation: The Missing Piece from Energy Duality
Mincheol Park, Heeji Won, Won Woo Ro, Suhyun Kim
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Re-examines entropy minimization in test-time adaptation through its dual relationship with energy-based likelihood, motivating an energy-aware TTA formulation.
Problem
Test-time adaptation under distribution shift commonly relies on entropy minimization, whose sufficiency for preserving performance is questioned.
Approach
Analyzes the dual relationship between entropy and energy, treating energy as a proxy for likelihood (lower energy means higher observability under the learned distribution) to show entropy minimization alone is insufficient and to motivate an energy-aware adaptation objective.
Experimental setup
Builds on: entropy minimization TTA
Closest papers by meaning
- Unified Entropy Optimization for Open-Set Test-Time Adaptation (CVPR 2024 · similarity 0.91)
- COME: Test-time Adaption by Conservatively Minimizing Entropy (ICLR 2025 · similarity 0.89)
- TEA: Test-time Energy Adaptation (CVPR 2024 · similarity 0.89)
- Beyond Entropy: Region Confidence Proxy for Wild Test-Time Adaptation (ICML 2025 · similarity 0.89)
- Backpropagation-Free Test-Time Adaptation via Probabilistic Gaussian Alignment (NeurIPS 2025 · similarity 0.88)
- Protected Test-Time Adaptation via Online Entropy Matching: A Betting Approach (NeurIPS 2024 · similarity 0.88)
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.
