FUSE adapts a multimodal diffusion transformer for flow-matching-based posterior estimation and adds inference-time likelihood-guided particle resampling, achieving state-of-the-art posterior fidelity on SBI benchmarks and real-world exoplanet orbital estimation.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
FUSE: FK-Steered Multi-Modal Flow Matching for Efficient Simulation-Based Posterior Estimation
FUSE adapts a multimodal diffusion transformer for flow-matching-based posterior estimation and adds inference-time likelihood-guided particle resampling, achieving state-of-the-art posterior fidelity on SBI benchmarks and real-world exoplanet orbital estimation.