# iGOS++: Integrated Gradient Optimized Saliency by Bilateral Perturbations

arXiv: https://arxiv.org/abs/2012.15783 · HTML: https://mlsanity.com/paper/2012.15783/

## Key facts

- Cited by 19 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 31 Dec 2020.

## Built on by (most cited first, 19 of 19)

- [Impossibility Theorems for Feature Attribution](https://mlsanity.com/paper/2212.11870/) (2022)
- [Cycle-Consistent Counterfactuals by Latent Transformations](https://mlsanity.com/paper/2203.15064/) (2022)
- [IDGI: A Framework to Eliminate Explanation Noise from Integrated Gradients](https://mlsanity.com/paper/2303.14242/) (2023)
- [Where do Large Vision-Language Models Look at when Answering Questions?](https://mlsanity.com/paper/2503.13891/) (2025)
- [Deletion and Insertion Tests in Regression Models](https://mlsanity.com/paper/2205.12423/) (2022)
- [What You See is What You Classify: Black Box Attributions](https://mlsanity.com/paper/2205.11266/) (2022)
- [Comparing the Decision-Making Mechanisms by Transformers and CNNs via Explanation Methods](https://mlsanity.com/paper/2212.06872/) (2022)
- [New Definitions and Evaluations for Saliency Methods: Staying Intrinsic, Complete and Sound](https://mlsanity.com/paper/2211.02912/) (2022)
- [From Heatmaps to Structural Explanations of Image Classifiers](https://mlsanity.com/paper/2109.06365/) (2021)
- [Computer Vision Methods for the Microstructural Analysis of Materials: The State-of-the-art and Future Perspectives](https://mlsanity.com/paper/2208.04149/) (2022)
- [Forward Learning for Gradient-based Black-box Saliency Map Generation](https://mlsanity.com/paper/2403.15603/) (2024)
- [Explainable Image Recognition via Enhanced Slot-attention Based Classifier](https://mlsanity.com/paper/2407.05616/) (2024)
- [Manifold-Aligned Guided Integrated Gradients for Reliable Feature Attribution](https://mlsanity.com/paper/2605.02167/) (2026)
- [Fidelity of Ensemble Aggregation for Saliency Map Explanations using Bayesian Optimization Techniques](https://mlsanity.com/paper/2207.01565/) (2022)
- [PhaseWin: An Efficient Search Algorithm for Faithful Visual Attribution](https://mlsanity.com/paper/2606.18008/) (2026)
- [Interactive Mars Image Content-Based Search with Interpretable Machine Learning](https://mlsanity.com/paper/2402.16860/) (2024)
- [Self-Supervised Learning for Building Robust Pediatric Chest X-ray Classification Models](https://mlsanity.com/paper/2409.00231/) (2024)
- [A Good Initialization is All You Need for Faithful Visual Attribution](https://mlsanity.com/paper/2607.06726/) (2026)
- [What Pixels Are Enough? SEAMS: Sufficiency Saliency via MSE-Preservation Soft-Masks](https://mlsanity.com/paper/2607.09164/) (2026)

## Abstract (arXiv)

The black-box nature of the deep networks makes the explanation for "why" they make certain predictions extremely challenging. Saliency maps are one of the most widely-used local explanation tools to alleviate this problem. One of the primary approaches for generating saliency maps is by optimizing a mask over the input dimensions so that the output of the network is influenced the most by the masking. However, prior work only studies such influence by removing evidence from the input. In this paper, we present iGOS++, a framework to generate saliency maps that are optimized for altering the output of the black-box system by either removing or preserving only a small fraction of the input. Additionally, we propose to add a bilateral total variation term to the optimization that improves the continuity of the saliency map especially under high resolution and with thin object parts. The evaluation results from comparing iGOS++ against state-of-the-art saliency map methods show significant improvement in locating salient regions that are directly interpretable by humans. We utilized iGOS++ in the task of classifying COVID-19 cases from x-ray images and discovered that sometimes the CNN network is overfitted to the characters printed on the x-ray images when performing classification. Fixing this issue by data cleansing significantly improved the precision and recall of the classifier.