用可证明的不确定性评估，让模型更聪明地选数据，少问人、多提效。
Annotation-Efficient Active Test-Time Adaptation with Conformal Prediction
- 基于校准预测的不确定性度量，确保每一步选择都有统计保障。
- 在多个数据集上比当前最佳方法提升约5%准确率。
- 适合需要高效利用人工标注资源的部署场景。
主动测试时自适应（ATTA）通过在部署时有选择地查询人工标注来提升模型在分布偏移下的鲁棒性，但现有方法依赖启发式不确定性度量，数据选取效率低，浪费标注预算。我们提出校准预测主动测试时自适应（CPATTA），首次将具有覆盖率保证的严格不确定性引入ATTA。CPATTA采用平滑的校准得分结合顶K确定性度量，基于伪覆盖率驱动的在线权重更新算法，适应人类监督的分布偏移检测器，以及分阶段更新策略以平衡人工标注与模型标注数据。大量实验表明，CPATTA在准确率上持续优于当前最佳ATTA方法约5%。代码与数据集已公开于 https://github.com/tingyushi/CPATTA。
原文摘要 · Abstract (English)
Active Test-Time Adaptation (ATTA) improves model robustness under domain shift by selectively querying human annotations at deployment, but existing methods use heuristic uncertainty measures and suffer from low data selection efficiency, wasting human annotation budget. We propose Conformal Prediction Active TTA (CPATTA), which first brings principled, coverage-guaranteed uncertainty into ATTA. CPATTA employs smoothed conformal scores with a top-K certainty measure, an online weight-update algorithm driven by pseudo coverage, a domain-shift detector that adapts human supervision, and a staged update scheme balances human-labeled and model-labeled data. Extensive experiments demonstrate that CPATTA consistently outperforms the state-of-the-art ATTA methods by around 5% in accuracy. Our code and datasets are available at https://github.com/tingyushi/CPATTA.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。