# Multivariate Scientific Data Compression with Learned Cross-Variable Latent Decorrelation and Autoregressive Entropy Modeling

arXiv: https://arxiv.org/abs/2608.30262 · HTML: https://mlsanity.com/paper/2608.30262/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 26 references that are in the map.
- First posted on arXiv 31 Aug 2026.

## Builds on (most cited first, 10 of 10)

- [Auto-Encoding Variational Bayes](https://mlsanity.com/paper/1312.6114/) (2013)
- [Context-adaptive Entropy Model for End-to-end Optimized Image Compression](https://mlsanity.com/paper/1809.10452/) (2018)
- [TTHRESH: Tensor Compression for Multidimensional Visual Data](https://mlsanity.com/paper/1806.05952/) (2018)
- [Compressing multidimensional weather and climate data into neural networks](https://mlsanity.com/paper/2210.12538/) (2022)
- [StelLA: Subspace Learning in Low-rank Adaptation using Stiefel Manifold](https://mlsanity.com/paper/2510.01938/) (2025)
- [A Survey on Error-Bounded Lossy Compression for Scientific Datasets](https://mlsanity.com/paper/2404.02840/) (2024)
- [MGARD: A multigrid framework for high-performance, error-controlled data compression and refactoring](https://mlsanity.com/paper/2401.05994/) (2024)
- [Parallel Tensor Compression for Large-Scale Scientific Data](https://mlsanity.com/paper/1510.06689/) (2015)
- [Generative Latent Diffusion for Efficient Spatiotemporal Data Reduction](https://mlsanity.com/paper/2507.02129/) (2025)
- [Residual Modeling for High-Fidelity Learned Compression of Scientific Data](https://mlsanity.com/paper/2606.05389/) (2026)

## Abstract (arXiv)

Scientific simulations generate collections of physical fields with heterogeneous statistics and dependencies, yet learned compressors often encode those fields independently or rely on a shared encoder without explicitly modeling the structure that remains in latent space. We present CAESAR-LDAR, an error-controlled multivariate learned compressor that augments a shared CAESAR-V backbone with two complementary mechanisms: a trainable orthogonal transform that reorganizes dependence across aligned latent channels, and a causal autoregressive hierarchical prior that captures local spatial structure left after transformation. Orthogonality is maintained through a matrix-exponential parameterization, making the transform exactly invertible without an additional penalty. A common residual-correction stage is applied uniformly to all variants to enforce the requested reconstruction tolerance. Experiments across combustion, climate, and turbulence data show that the two mechanisms are useful in different regimes. Latent decorrelation helps most when substantial linear cross-channel dependence survives the nonlinear encoder, whereas autoregressive modeling remains effective when the remaining structure is primarily local or spatial. Their combination provides the strongest or near-strongest rate-distortion performance across the evaluated datasets. The global transform adds little computational overhead, while autoregressive coding introduces a larger throughput tradeoff. More broadly, the results suggest a practical design principle for multivariate scientific compression: exploit global cross-channel dependence when it is measurably present in latent space, and use local probabilistic context as a complementary mechanism across a wider range of data regimes.