Latent Autoregressive Source Separation

Citations Over Time

Abstract

Autoregressive models have achieved impressive results over a wide range of domains in terms of generation quality and downstream task performance. In the continuous domain, a key factor behind this success is the usage of quantized latent spaces (e.g., obtained via VQ-VAE autoencoders), which allow for dimensionality reduction and faster inference times. However, using existing pre-trained models to perform new non-trivial tasks is difficult since it requires additional fine-tuning or extensive training to elicit prompting. This paper introduces LASS as a way to perform vector-quantized Latent Autoregressive Source Separation (i.e., de-mixing an input signal into its constituent sources) without requiring additional gradient-based optimization or modifications of existing models. Our separation method relies on the Bayesian formulation in which the autoregressive models are the priors, and a discrete (non-parametric) likelihood function is constructed by performing frequency counts over latent sums of addend tokens. We test our method on images and audio with several sampling strategies (e.g., ancestral, beam search) showing competitive results with existing approaches in terms of separation quality while offering at the same time significant speedups in terms of inference time and scalability to higher dimensional data.

Related Papers

• → Model Parameterization, Prior Distributions, and the General Time-Reversible Model in Bayesian Phylogenetics(2004)95 cited
• → Applications of Bayesian Methods to Epidemiologic Research(2014)38 cited
• → The use of local and nonlocal priors in Bayesian test‐based monitoring for single‐arm phase II clinical trials(2021)4 cited
• → Bias and the Effect of Priors in Bayesian Estimation of Parameters of Item Response Models(1990)38 cited
• → Informative priors or noninformative priors? A Bayesian re-analysis of binary data from Macugen phase III clinical trials(2016)4 cited