# Apple 提出 GH-ESD：面向实例级视觉任务的假设驱动错误切片发现方法

- 来源：Apple Machine Learning Research（RSS）
- 发布时间：2026-07-27 08:00
- AIHOT 分数：40
- AIHOT 链接：https://aihot.virxact.com/items/cms3el9cc000orozqcae978wd
- 原文链接：https://machinelearning.apple.com/research/gh-esd

## AI 摘要

Apple 机器学习研究团队提出 GH-ESD（Grounded Hypothesis-Driven Error Slice Discovery），一种针对目标检测与分割等实例级视觉任务的错误切片发现方法。现有方法主要适用于图像级分类，难以捕捉由上下文关系和空间视觉模式导致的实例级失败。GH-ESD 通过基于假设的驱动方式，系统性地发现模型在语义连贯子集上的系统性失效。

## 正文

Systematic failures of vision models on semantically coherent subsets, known as error slices, reveal limitations in robustness and evaluation. Existing slice discovery approaches largely model slices as clusters in representation space or combinations of predefined attributes. While effective for image-level classification, such formulations are insufficient for instance-level tasks such as object detection and segmentation, where failures often arise from contextual relational and spatially grounded visual patterns. We propose GH-ESD (Grounded Hypothesis-Driven Error Slice Discovery), a…