# Evidential Sparsification of Multimodal Latent Spaces in Conditional Variational Autoencoders

arXiv: https://arxiv.org/abs/2010.09164 · HTML: https://mlsanity.com/paper/2010.09164/

## Key facts

- Cited by 15 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Oct 2020.

## Built on by (most cited first, 15 of 15)

- [Foundations and Trends in Multimodal Machine Learning: Principles, Challenges, and Open Questions](https://mlsanity.com/paper/2209.03430/) (2022)
- [Interpretable Self-Aware Neural Networks for Robust Trajectory Prediction](https://mlsanity.com/paper/2211.08701/) (2022)
- [Heterogeneous-Agent Trajectory Forecasting Incorporating Class Uncertainty](https://mlsanity.com/paper/2104.12446/) (2021)
- [Random-Set Neural Networks (RS-NN)](https://mlsanity.com/paper/2307.05772/) (2023)
- [Propagating State Uncertainty Through Trajectory Forecasting](https://mlsanity.com/paper/2110.03267/) (2021)
- [Exploring Social Posterior Collapse in Variational Autoencoder for Interaction Modeling](https://mlsanity.com/paper/2112.00298/) (2021)
- [Multi-Agent Variational Occlusion Inference Using People as Sensors](https://mlsanity.com/paper/2109.02173/) (2021)
- [DySurv: dynamic deep learning model for survival analysis with conditional variational inference](https://mlsanity.com/paper/2310.18681/) (2023)
- [Evidential Softmax for Sparse Multimodal Distributions in Deep Generative Models](https://mlsanity.com/paper/2110.14182/) (2021)
- [MultiMax: Sparse and Multi-Modal Attention Learning](https://mlsanity.com/paper/2406.01189/) (2024)
- [ELVAE: Evidential Learning-Based Variational Autoencoder for Uncertainty-Aware Generation](https://mlsanity.com/paper/2608.10398/) (2026)
- [Sparse Communication via Mixed Distributions](https://mlsanity.com/paper/2108.02658/) (2021)
- [Learning Discrete Structured Variational Auto-Encoder using Natural Evolution Strategies](https://mlsanity.com/paper/2205.01324/) (2022)
- [Foundations of Multisensory Artificial Intelligence](https://mlsanity.com/paper/2404.18976/) (2024)
- [When Should Graph Attention Be Sparse? Learning a Per-Edge Tsallis Index](https://mlsanity.com/paper/2608.02938/) (2026)

## Abstract (arXiv)

Discrete latent spaces in variational autoencoders have been shown to effectively capture the data distribution for many real-world problems such as natural language understanding, human intent prediction, and visual scene representation. However, discrete latent spaces need to be sufficiently large to capture the complexities of real-world data, rendering downstream tasks computationally challenging. For instance, performing motion planning in a high-dimensional latent representation of the environment could be intractable. We consider the problem of sparsifying the discrete latent space of a trained conditional variational autoencoder, while preserving its learned multimodality. As a post hoc latent space reduction technique, we use evidential theory to identify the latent classes that receive direct evidence from a particular input condition and filter out those that do not. Experiments on diverse tasks, such as image generation and human behavior prediction, demonstrate the effectiveness of our proposed technique at reducing the discrete latent sample space size of a model while maintaining its learned multimodality.