# ShapKO: Shapley-Adaptive Modality Knockout for Robust Multimodal Learning

arXiv: https://arxiv.org/abs/2607.09884 · HTML: https://mlsanity.com/paper/2607.09884/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 18 of its 28 references that are in the map.
- First posted on arXiv 10 Jul 2026.

## Builds on (most cited first, 18 of 18)

- [Balanced Multimodal Learning via On-the-fly Gradient Modulation](https://mlsanity.com/paper/2203.15332/) (2022)
- [Multimodal Generative Models for Scalable Weakly-Supervised Learning](https://mlsanity.com/paper/1802.05335/) (2018)
- [HeMIS: Hetero-Modal Image Segmentation](https://mlsanity.com/paper/1607.05194/) (2016)
- [Sampling Permutations for Shapley Value Estimation](https://mlsanity.com/paper/2104.12199/) (2021)
- [Deep Multimodal Learning with Missing Modality: A Survey](https://mlsanity.com/paper/2409.07825/) (2024)
- [Multimodal Representation Learning by Alternating Unimodal Adaptation](https://mlsanity.com/paper/2311.10707/) (2023)
- [Boosting Multi-modal Model Performance with Adaptive Gradient Modulation](https://mlsanity.com/paper/2308.07686/) (2023)
- [Hetero-Modal Variational Encoder-Decoder for Joint Modality Completion and Segmentation](https://mlsanity.com/paper/1907.11150/) (2019)
- [Classifier-guided Gradient Modulation for Enhanced Multimodal Learning](https://mlsanity.com/paper/2411.01409/) (2024)
- [Mitigating Modality Collapse in Multimodal VAEs via Impartial Optimization](https://mlsanity.com/paper/2206.04496/) (2022)
- [Survival Prediction of Brain Cancer with Incomplete Radiology, Pathology, Genomics, and Demographic Data](https://mlsanity.com/paper/2203.04419/) (2022)
- [Regression-adjusted Monte Carlo Estimators for Shapley Values and Probabilistic Values](https://mlsanity.com/paper/2506.11849/) (2025)
- [Learning End-to-end Multimodal Sensor Policies for Autonomous Navigation](https://mlsanity.com/paper/1705.10422/) (2017)
- [Improving Multimodal Learning with Multi-Loss Gradient Modulation](https://mlsanity.com/paper/2405.07930/) (2024)
- [FlexCare: Leveraging Cross-Task Synergy for Flexible Multimodal Healthcare Prediction](https://mlsanity.com/paper/2406.11928/) (2024)
- [What Makes for Robust Multi-Modal Models in the Face of Missing Modalities?](https://mlsanity.com/paper/2310.06383/) (2023)
- [Knockout: A simple way to handle missing inputs](https://mlsanity.com/paper/2405.20448/) (2024)
- [Investigating the Effect of Sensor Modalities in Multi-Sensor Detection-Prediction Models](https://mlsanity.com/paper/2101.03279/) (2021)

## Abstract (arXiv)

Multimodal medical models often degrade when inputs are missing, a common scenario in real-world clinical workflows. Separately, even when all modalities are present, modality dominance is observed during training, where optimization over-relies on a highly predictive modality and undertrains complementary sources, resulting in poor robustness under partial availability. While training-time modality knockout improves missing-modality robustness, existing approaches use static masking rates that cannot adapt to evolving modality utility during training. We introduce ShapKO (Shapley-Adaptive Modality Knockout), a dynamic training strategy that learns modality-specific knockout probabilities based on validation utility. ShapKO periodically evaluates performance across modality subsets, estimates modality importance via Shapley values, and updates masking probabilities to suppress dominant modalities more frequently. This adaptive process promotes complementary representations, while requiring no architectural modifications. We evaluate ShapKO on three datasets covering multitask clinical classification, survival prediction, and cancer detection. ShapKO consistently improves performance under modality absence and yields interpretable trajectories of learned masking behavior. Code is available at: https://github.com/sumona00/ShapKO