[
    {
        "id": "osp-22611",
        "type": "article-journal",
        "title": "E-MoE: Enhanced Mixture-of-Experts for Non-Factorized Diffusion Language Models",
        "author": [
            {
                "family": "Ivanov",
                "given": "Arseny"
            },
            {
                "family": "Kolesov",
                "given": "Alexander"
            },
            {
                "family": "Korotin",
                "given": "Alexander"
            },
            {
                "family": "Oseledets",
                "given": "Ivan"
            },
            {
                "family": "Goncharov",
                "given": "Mikhail"
            }
        ],
        "URL": "https://omanscience.com/ar/articles/e-moe-enhanced-mixture-of-experts-for-non-factorized-diffusion-language-models",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Masked diffusion models (MDMs) generate sequences by progressively unmasking several tokens per denoising step, but their reverse process is typically factorized over positions, limiting sample quality in the few-step regime where diffusion's speed advantage over autoregressive decoding matters most. A recent line of work introduces a continuous Gaussian latent, trained as a variational autoencoder, to capture correlations across positions, but such approaches are prone to posterior collapse, where the latent is silently ignored. We propose Enhanced Mixture-of-Experts (E-MoE), which builds the reverse process as a mixture of factorized distributions over a discrete shared latent given by the expert-routing decisions of a Mixture-of-Experts (MoE) backbone, without increasing active parameters over the factorized baseline. Across synthetic multi-modal benchmarks, binarized MNIST, and LM1B, E-MoE improves few-step generation over factorized baselines."
    }
]