# CHILD: Human-in-the-Loop OOD Detection for Safe Clinical Deployment

arXiv: https://arxiv.org/abs/2609.07188 · HTML: https://mlsanity.com/paper/2609.07188/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 32 references that are in the map.
- First posted on arXiv 7 Sep 2026.

## Builds on (most cited first, 21 of 21)

- [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](https://mlsanity.com/paper/1610.02136/) (2016)
- [Energy-based Out-of-distribution Detection](https://mlsanity.com/paper/2010.03759/) (2020)
- [Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images](https://mlsanity.com/paper/1412.1897/) (2014)
- [ReAct: Out-of-distribution Detection With Rectified Activations](https://mlsanity.com/paper/2111.12797/) (2021)
- [Scaling Out-of-Distribution Detection for Real-World Settings](https://mlsanity.com/paper/1911.11132/) (2019)
- [Extremely Simple Activation Shaping for Out-of-Distribution Detection](https://mlsanity.com/paper/2209.09858/) (2022)
- [LoCoOp: Few-Shot Out-of-Distribution Detection via Prompt Learning](https://mlsanity.com/paper/2306.01293/) (2023)
- [Historical Test-time Prompt Tuning for Vision Foundation Models](https://mlsanity.com/paper/2410.20346/) (2024)
- [OODD: Test-time Out-of-Distribution Detection with Dynamic Dictionary](https://mlsanity.com/paper/2503.10468/) (2025)
- [FodFoM: Fake Outlier Data by Foundation Models Creates Stronger Visual Out-of-Distribution Detector](https://mlsanity.com/paper/2412.05293/) (2024)
- [Noisy Test-Time Adaptation in Vision-Language Models](https://mlsanity.com/paper/2502.14604/) (2025)
- [Model-free Test Time Adaptation for Out-Of-Distribution Detection](https://mlsanity.com/paper/2311.16420/) (2023)
- [CADRef: Robust Out-of-Distribution Detection via Class-Aware Decoupled Relative Feature Leveraging](https://mlsanity.com/paper/2503.00325/) (2025)
- [Are We Ready for Out-of-Distribution Detection in Digital Pathology?](https://mlsanity.com/paper/2407.13708/) (2024)
- [TagFog: Textual Anchor Guidance and Fake Outlier Generation for Visual Out-of-Distribution Detection](https://mlsanity.com/paper/2412.05292/) (2024)
- [Out-of-Distribution Learning with Human Feedback](https://mlsanity.com/paper/2408.07772/) (2024)
- [Revisiting Logit Distributions for Reliable Out-of-Distribution Detection](https://mlsanity.com/paper/2510.20134/) (2025)
- [FA: Forced Prompt Learning of Vision-Language Models for Out-of-Distribution Detection](https://mlsanity.com/paper/2507.04511/) (2025)
- [DCAC: Dynamic Class-Aware Cache Creates Stronger Out-of-Distribution Detectors](https://mlsanity.com/paper/2601.12468/) (2026)
- [Hierarchical Vision-Language Learning for Medical Out-of-Distribution Detection](https://mlsanity.com/paper/2508.17667/) (2025)
- [TTL: Test-time Textual Learning for OOD Detection with Pretrained Vision-Language Models](https://mlsanity.com/paper/2604.15756/) (2026)

## Abstract (arXiv)

Out-of-distribution (OOD) detection is critical for safe deployment of medical AI systems. Recently, test-time adaptation (TTA) has emerged as a new paradigm for OOD detection, automatically adjusting detector behavior during deployment. However, such automatic adaptation mechanisms may raise safety concerns in safety-critical clinical environments. While physician oversight can mitigate these risks, it is resource-intensive and must be judiciously allocated. To reconcile safety with efficiency, we propose CHILD, a training-free framework designed to enhance streaming OOD detection via sparse human feedback. Operating under strict budget constraints, CHILD employs an adaptive risk-aware sample selection mechanism to pinpoint only the most decision-uncertain samples for review. Crucially, it maximizes the utility of this sparse feedback through a retrieval-based score calibration module, which refines model predictions using a compact feature cache without any parameter updates. Extensive experiments on four medical benchmarks demonstrate that CHILD turns limited supervision into significant reliability gains: with a sparse feedback budget of only 5%, it reduces the average FPR95 from 72.63% to 60.26% and improves AUROC from 75.53% to 81.85%, consistently outperforming state-of-the-art baselines. Our code is publicly available at https://github.com/figec/CHILD.