SliceLens: Fine-Grained and Grounded Error Slice Discovery for Multi-Instance Vision Tasks
This addresses the challenge of robust model evaluation for computer vision practitioners by enabling interpretable error analysis in complex multi-instance tasks, though it builds incrementally on existing slice discovery concepts.
The paper tackles the problem of identifying systematic failures (error slices) in multi-instance vision tasks like detection and segmentation, where existing methods are limited to image classification. It proposes SliceLens, a framework using LLMs and VLMs for fine-grained error slice discovery, achieving state-of-the-art performance with a Precision@10 improvement of 0.42 (0.73 vs. 0.31) on a new benchmark.
Systematic failures of computer vision models on subsets with coherent visual patterns, known as error slices, pose a critical challenge for robust model evaluation. Existing slice discovery methods are primarily developed for image classification, limiting their applicability to multi-instance tasks such as detection, segmentation, and pose estimation. In real-world scenarios, error slices often arise from corner cases involving complex visual relationships, where existing instance-level approaches lacking fine-grained reasoning struggle to yield meaningful insights. Moreover, current benchmarks are typically tailored to specific algorithms or biased toward image classification, with artificial ground truth that fails to reflect real model failures. To address these limitations, we propose SliceLens, a hypothesis-driven framework that leverages LLMs and VLMs to generate and verify diverse failure hypotheses through grounded visual reasoning, enabling reliable identification of fine-grained and interpretable error slices. We further introduce FeSD (Fine-grained Slice Discovery), the first benchmark specifically designed for evaluating fine-grained error slice discovery across instance-level vision tasks, featuring expert-annotated and carefully refined ground-truth slices with precise grounding to local error regions. Extensive experiments on both existing benchmarks and FeSD demonstrate that SliceLens achieves state-of-the-art performance, improving Precision@10 by 0.42 (0.73 vs. 0.31) on FeSD, and identifies interpretable slices that facilitate actionable model improvements, as validated through model repair experiments.