Research map / Label-efficient and robust learning / Test-time adaptation
STAR: Test-Time Adaptation Can Enhance Universal Prompt Learning for Vision-Language Models
Yiwei Fu, Hui Wan, Xiao Luo, Minghua Deng
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 test-time adaptation (M3) on top of prompt-tuned model (ablation) | STAR_L · iNaturalist+SUN average | FPR95: 38.36 -> 14.55 | Improved |
| use Fisher-score adaptive soft gating instead of fixed threshold 0.6 log K (ablation) | STAR_L · iNaturalist+SUN average | FPR95: 83.02 -> 14.55 | Improved |
| use conjugate uncertainty-aware loss instead of softmax entropy minimization (ablation) | STAR_L · iNaturalist+SUN average | FPR95: 87.24 -> 14.55 | Improved |
| add STAR test-time adaptation to 1-shot LoCoOp | LoCoOp_G · ImageNet-1k ID, OOD average (1-shot) | FPR95: 33.08 -> 27.03 | Improved |
| add STAR test-time adaptation to 16-shot SCT | SCT_G · ImageNet-1k ID, OOD average (16-shot) | FPR95: 26.62 -> 23.17 | Improved |
| add STAR to LoCoOp on Texture | LoCoOp_G (1-shot) · Texture (1-shot) | FPR95: 50.37 -> 52.25 | Hurt |
| add STAR on iNaturalist harmonic accuracy | LoCoOp (1-shot) · iNaturalist (1-shot) | harmonic accuracy ACC_H: 42.72 -> 41.36 | Hurt |
| add STAR on harmonic accuracy average | LoCoOp (16-shot) · ImageNet-1k ID, OOD average (16-shot) | harmonic accuracy ACC_H: 39.85 -> 46.04 | Improved |
| decrease adaptation batch size from 512 to 32 (ablation) | STAR · iNaturalist | FPR95 | Improved |
| change batch size (ablation) | STAR · SUN | FPR95 | No clear effect |
| switch image encoder from ViT-B/16 to ViT-B/32 (ablation) | STAR · iNaturalist and SUN | FPR95 / AUROC | No clear effect |
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.
- LoCoOp — LoCoOp: Few-Shot Out-of-Distribution Detection via Prompt Learning (NeurIPS 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
- Frustratingly Easy Test-Time Adaptation of Vision-Language Models (NeurIPS 2024 · similarity 0.91)
- Black-Box Test-Time Prompt Tuning for Vision-Language Models (AAAI 2025 · similarity 0.90)
- The Illusion of Progress? A Critical Look at Test-Time Adaptation for Vision-Language Models (NeurIPS 2025 · similarity 0.90)
- Dual Prototype Evolving for Test-Time Generalization of Vision-Language Models (NeurIPS 2024 · similarity 0.89)
- Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts (NeurIPS 2026 · similarity 0.89)
- BoostAdapter: Improving Vision-Language Test-Time Adaptation via Regional Bootstrapping (NeurIPS 2024 · similarity 0.89)
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.
