The increasing realism of synthetic speech, driven by advancements in text-tospeech models, raises ethical concerns regarding impersonation and disinformation. Audio watermarking offers a promising solution via embedding humanimperceptible watermarks into AI-generated audios. However, the robustness of audio watermarking against common/adversarial perturbations remains understudied.

The execution time for inference is not provided in the paper. We will state this in the next revision. The advantage of the proposed algorithm is clear for the discrete tasks but not for continuous tasks. The results are competitive with SOTA so we have elected to include them for completeness. I think the authors used seven datasets out of the eight datasets described in the paper [7].

This often requires maximizing the likelihood of a symbolic constraint w.r.t. the neural network's output distribution. Such output distributions are typically assumed to be fully-factorized. This limits the applicability of neuro-symbolic learning to the more expressive autoregressive distributions, e.g., transformers. Under such distributions, computing the likelihood of even simple constraints is #P-hard. Instead of attempting to enforce the constraint on the entire output distribution, we propose to do so on a random, local approximation thereof.

We consider the problem of training machine learning models on distributed data in a decentralized way. For finite-sum problems, fast single-machine algorithms for large datasets rely on stochastic updates combined with variance reduction. Yet, existing decentralized stochastic algorithms either do not obtain the full speedup allowed by stochastic updates, or require oracles that are more expensive than regular gradients. In this work, we introduce a Decentralized stochastic algorithm with Variance Reduction called DVR. DVR only requires computing stochastic gradients of the local functions, and is computationally as fast as a standard stochastic variance-reduced algorithms run on a 1/n fraction of the dataset, where n is the number of nodes. To derive DVR, we use Bregman coordinate descent on a well-chosen dual problem, and obtain a dual-free algorithm using a specific Bregman divergence.