13 citations · 31 across the 16 of their papers we have counts for
12 papers · 1 filter
The Key to Going Linear: Analysis-Driven Transformer Linearization
Anna Kuzina, Paul N. Whatmough, Babak Ehteshami Bejnordi
The quadratic cost of causal self-attention severely bottlenecks long-context transformer inference. While numerous post hoc linearization pipelines exist, it is difficult to ident…
Efficient Reasoning on the Edge
Yelysei Bondarenko, Thomas Hehn, Rob Hesselink +15
Large language models (LLMs) with chain-of-thought reasoning achieve state-of-the-art performance across complex problem-solving tasks, but their verbose reasoning traces and large…
KaVa: Latent Reasoning via Compressed KV-Cache Distillation
Anna Kuzina, Maciej Pioro, Paul N. Whatmough +1
Large Language Models (LLMs) excel at multi-step reasoning problems with explicit chain-of-thought (CoT), but verbose traces incur significant computational costs and memory overhe…
Hierarchical VAE with a Diffusion-based VampPrior
Anna Kuzina, Jakub M. Tomczak
Deep hierarchical variational autoencoders (VAEs) are powerful latent variable generative models. In this paper, we introduce Hierarchical VAE with Diffusion-based Variational Mixt…
Variational Stochastic Gradient Descent for Deep Neural Networks
Haotian Chen, Anna Kuzina, Babak Esmaeili +1
Current state-of-the-art optimizers are adaptive gradient-based optimization methods such as Adam. Recently, there has been an increasing interest in formulating gradient-based opt…
Exploring Continual Learning of Diffusion Models
Michał Zając, Kamil Deja, Anna Kuzina +4
Diffusion models have achieved remarkable success in generating high-quality images thanks to their novel training procedures applied to unprecedented amounts of data. However, tra…
Discouraging posterior collapse in hierarchical Variational Autoencoders using context
Anna Kuzina, Jakub M. Tomczak
Hierarchical Variational Autoencoders (VAEs) are among the most popular likelihood-based generative models. There is a consensus that the top-down hierarchical VAEs allow effective…
Equivariant Priors for Compressed Sensing with Unknown Orientation
Anna Kuzina, Kumar Pratik, Fabio Valerio Massoli +1
In compressed sensing, the goal is to reconstruct the signal from an underdetermined system of linear measurements. Thus, prior knowledge about the signal of interest and its struc…
On Analyzing Generative and Denoising Capabilities of Diffusion-based Deep Generative Models
Kamil Deja, Anna Kuzina, Tomasz Trzciński +1
Diffusion-based Deep Generative Models (DDGMs) offer state-of-the-art performance in generative modeling. Their main strength comes from their unique setup in which a model (the ba…
Alleviating Adversarial Attacks on Variational Autoencoders with MCMC
Anna Kuzina, Max Welling, Jakub M. Tomczak
Variational autoencoders (VAEs) are latent variable models that can generate complex objects and provide meaningful latent representations. Moreover, they could be further used in…
CKConv: Continuous Kernel Convolution For Sequential Data
David W. Romero, Anna Kuzina, Erik J. Bekkers +2
Conventional neural architectures for sequential data present important limitations. Recurrent networks suffer from exploding and vanishing gradients, small effective memory horizo…
BooVAE: Boosting Approach for Continual Learning of VAE
Anna Kuzina, Evgenii Egorov, Evgeny Burnaev
Variational autoencoder (VAE) is a deep generative model for unsupervised learning, allowing to encode observations into the meaningful latent space. VAE is prone to catastrophic f…