arXiv subjects

Jan Chorowski

Publications and source records attributed to Jan Chorowski.

At least 19 recordsLinked to original sources

arXiv2026

BDH-CQ: In-Context Learning with Recurrent Latent Reasoning

We introduce BDH-CQ, a reasoning model that combines in-context learning with recurrent latent reasoning. Inputs presented at inference time continuously update the model's recurrent memory; the model then solves a query through iterative computation in a high-dimensional latent space, without verbalizing its intermediate reasoning. We evaluate the model on the public ARC-AGI-1 evaluation set and use controlled ARC-like interventions to study what it learns from demonstrations, how consistently it applies an inferred transformation, and which concepts remain difficult. A 150M-parameter configuration reaches 29.5% pass@2 at a computed inference cost of \$0.0007 per task. This operating point breaks through the previously reported ARC-AGI-1 cost-accuracy Pareto frontier, establishing a new state of the art in benchmark cost efficiency.

cs.NE↗

arXiv2025

The Dragon Hatchling: The Missing Link between the Transformer and Models of the Brain

The relationship between computing systems and the brain has served as motivation for pioneering theoreticians since John von Neumann and Alan Turing. Uniform, scale-free biological networks, such as the brain, have powerful properties, including generalizing over time, which is the main barrier for Machine Learning on the path to Universal Reasoning Models. We introduce `Dragon Hatchling' (BDH), a new Large Language Model architecture based on a scale-free biologically inspired network of \$n\$ locally-interacting neuron particles. BDH couples strong theoretical foundations and inherent interpretability without sacrificing Transformer-like performance. BDH is a practical, performant state-of-the-art attention-based state space sequence learning architecture. In addition to being a graph model, BDH admits a GPU-friendly formulation. It exhibits Transformer-like scaling laws: empirically BDH rivals GPT2 performance on language and translation tasks, at the same number of parameters (10M to 1B), for the same training data. BDH can be represented as a brain model. The working memory of BDH during inference entirely relies on synaptic plasticity with Hebbian learning using spiking neurons. We confirm empirically that specific, individual synapses strengthen connection whenever BDH hears or reasons about a specific concept while processing language inputs. The neuron interaction network of BDH is a graph of high modularity with heavy-tailed degree distribution. The BDH model is biologically plausible, explaining one possible mechanism which human neurons could use to achieve speech. BDH is designed for interpretability. Activation vectors of BDH are sparse and positive. We demonstrate monosemanticity in BDH on language tasks. Interpretability of state, which goes beyond interpretability of neurons and model parameters, is an inherent feature of the BDH architecture.

cs.NE↗

arXiv2023

Pathway: a fast and flexible unified stream data processing framework for analytical and Machine Learning applications

We present Pathway, a new unified data processing framework that can run workloads on both bounded and unbounded data streams. The framework was created with the original motivation of resolving challenges faced when analyzing and processing data from the physical economy, including streams of data generated by IoT and enterprise systems. These required rapid reaction while calling for the application of advanced computation paradigms (machinelearning-powered analytics, contextual analysis, and other elements of complex event processing). Pathway is equipped with a Table API tailored for Python and Python/SQL workflows, and is powered by a distributed incremental dataflow in Rust. We describe the system and present benchmarking results which demonstrate its capabilities in both batch and streaming contexts, where it is able to surpass state-of-the-art industry frameworks in both scenarios. We also discuss streaming use cases handled by Pathway which cannot be easily resolved with state-of-the-art industry frameworks, such as streaming iterative graph algorithms (PageRank, etc.).

cs.LG↗

arXiv2022

Efficient Transformers with Dynamic Token Pooling

Transformers achieve unrivalled performance in modelling language, but remain inefficient in terms of memory and time complexity. A possible remedy is to reduce the sequence length in the intermediate layers by pooling fixed-length segments of tokens. Nevertheless, natural units of meaning, such as words or phrases, display varying sizes. To address this mismatch, we equip language models with a dynamic-pooling mechanism, which predicts segment boundaries in an autoregressive fashion. We compare several methods to infer boundaries, including end-to-end learning through stochastic re-parameterisation, supervised learning (based on segmentations from subword tokenizers or spikes in conditional entropy), as well as linguistically motivated boundaries. We perform character-level evaluation on texts from multiple datasets and morphologically diverse languages. The results demonstrate that dynamic pooling, which jointly segments and models language, is both faster and more accurate than vanilla Transformers and fixed-length pooling within the same computational budget.

cs.CL↗

arXiv2022

Variable-rate hierarchical CPC leads to acoustic unit discovery in speech

The success of deep learning comes from its ability to capture the hierarchical structure of data by learning high-level representations defined in terms of low-level ones. In this paper we explore self-supervised learning of hierarchical representations of speech by applying multiple levels of Contrastive Predictive Coding (CPC). We observe that simply stacking two CPC models does not yield significant improvements over single-level architectures. Inspired by the fact that speech is often described as a sequence of discrete units unevenly distributed in time, we propose a model in which the output of a low-level CPC module is non-uniformly downsampled to directly minimize the loss of a high-level CPC module. The latter is designed to also enforce a prior of separability and discreteness in its representations by enforcing dissimilarity of successive high-level representations through focused negative sampling, and by quantization of the prediction targets. Accounting for the structure of the speech signal improves upon single-level CPC features and enhances the disentanglement of the learned representations, as measured by downstream speech recognition tasks, while resulting in a meaningful segmentation of the signal that closely resembles phone boundaries.

cs.SD↗

arXiv2021

Contrastive prediction strategies for unsupervised segmentation and categorization of phonemes and words

We investigate the performance on phoneme categorization and phoneme and word segmentation of several self-supervised learning (SSL) methods based on Contrastive Predictive Coding (CPC). Our experiments show that with the existing algorithms there is a trade off between categorization and segmentation performance. We investigate the source of this conflict and conclude that the use of context building networks, albeit necessary for superior performance on categorization tasks, harms segmentation performance by causing a temporal shift on the learned representations. Aiming to bridge this gap, we take inspiration from the leading approach on segmentation, which simultaneously models the speech signal at the frame and phoneme level, and incorporate multi-level modelling into Aligned CPC (ACPC), a variation of CPC which exhibits the best performance on categorization tasks. Our multi-level ACPC (mACPC) improves in all categorization metrics and achieves state-of-the-art performance in word segmentation.

cs.LG↗

arXiv2021

Information Retrieval for ZeroSpeech 2021: The Submission by University of Wroclaw

We present a number of low-resource approaches to the tasks of the Zero Resource Speech Challenge 2021. We build on the unsupervised representations of speech proposed by the organizers as a baseline, derived from CPC and clustered with the k-means algorithm. We demonstrate that simple methods of refining those representations can narrow the gap, or even improve upon the solutions which use a high computational budget. The results lead to the conclusion that the CPC-derived representations are still too noisy for training language models, but stable enough for simpler forms of pattern matching and retrieval.

cs.LG↗

arXiv2021

Aligned Contrastive Predictive Coding

We investigate the possibility of forcing a self-supervised model trained using a contrastive predictive loss to extract slowly varying latent representations. Rather than producing individual predictions for each of the future representations, the model emits a sequence of predictions shorter than that of the upcoming representations to which they will be aligned. In this way, the prediction network solves a simpler task of predicting the next symbols, but not their exact timing, while the encoding network is trained to produce piece-wise constant latent codes. We evaluate the model on a speech coding task and demonstrate that the proposed Aligned Contrastive Predictive Coding (ACPC) leads to higher linear phone prediction accuracy and lower ABX error rates, while being slightly faster to train due to the reduced number of prediction heads.

cs.LG↗

arXiv2020

Representing Point Clouds with Generative Conditional Invertible Flow Networks

In this paper, we propose a simple yet effective method to represent point clouds as sets of samples drawn from a cloud-specific probability distribution. This interpretation matches intrinsic characteristics of point clouds: the number of points and their ordering within a cloud is not important as all points are drawn from the proximity of the object boundary. We postulate to represent each cloud as a parameterized probability distribution defined by a generative neural network. Once trained, such a model provides a natural framework for point cloud manipulation operations, such as aligning a new cloud into a default spatial orientation. To exploit similarities between same-class objects and to improve model performance, we turn to weight sharing: networks that model densities of points belonging to objects in the same family share all parameters with the exception of a small, object-specific embedding vector. We show that these embedding vectors capture semantic relationships between objects. Our method leverages generative invertible flow networks to learn embeddings as well as to generate point clouds. Thanks to this formulation and contrary to similar approaches, we are able to train our model in an end-to-end fashion. As a result, our model offers competitive or superior quantitative results on benchmark datasets, while enabling unprecedented capabilities to perform cloud manipulation tasks, such as point cloud registration and regeneration, by a generative network.

cs.CV↗

arXiv2020

A Convolutional Deep Markov Model for Unsupervised Speech Representation Learning

Probabilistic Latent Variable Models (LVMs) provide an alternative to self-supervised learning approaches for linguistic representation learning from speech. LVMs admit an intuitive probabilistic interpretation where the latent structure shapes the information extracted from the signal. Even though LVMs have recently seen a renewed interest due to the introduction of Variational Autoencoders (VAEs), their use for speech representation learning remains largely unexplored. In this work, we propose Convolutional Deep Markov Model (ConvDMM), a Gaussian state-space model with non-linear emission and transition functions modelled by deep neural networks. This unsupervised model is trained using black box variational inference. A deep convolutional neural network is used as an inference network for structured variational approximation. When trained on a large scale speech dataset (LibriSpeech), ConvDMM produces features that significantly outperform multiple self-supervised feature extracting methods on linear phone classification and recognition on the Wall Street Journal dataset. Furthermore, we found that ConvDMM complements self-supervised methods like Wav2Vec and PASE, improving on the results achieved with any of the methods alone. Lastly, we find that ConvDMM features enable learning better phone recognizers than any other features in an extreme low-resource regime with few labeled training examples.

eess.AS↗

arXiv2020

Robust Training of Vector Quantized Bottleneck Models

In this paper we demonstrate methods for reliable and efficient training of discrete representation using Vector-Quantized Variational Auto-Encoder models (VQ-VAEs). Discrete latent variable models have been shown to learn nontrivial representations of speech, applicable to unsupervised voice conversion and reaching state-of-the-art performance on unit discovery tasks. For unsupervised representation learning, they became viable alternatives to continuous latent variable models such as the Variational Auto-Encoder (VAE). However, training deep discrete variable models is challenging, due to the inherent non-differentiability of the discretization operation. In this paper we focus on VQ-VAE, a state-of-the-art discrete bottleneck model shown to perform on par with its continuous counterparts. It quantizes encoder outputs with on-line $k$-means clustering. We show that the codebook learning can suffer from poor initialization and non-stationarity of clustered encoder outputs. We demonstrate that these can be successfully overcome by increasing the learning rate for the codebook and periodic date-dependent codeword re-initialization. As a result, we achieve more robust training across different tasks, and significantly increase the usage of latent codewords even for large codebooks. This has practical benefit, for instance, in unsupervised representation learning, where large codebooks may lead to disentanglement of latent representations.

cs.LG↗

arXiv2019

Conditional Invertible Flow for Point Cloud Generation

This paper focuses on a novel generative approach for 3D point clouds that makes use of invertible flow-based models. The main idea of the method is to treat a point cloud as a probability density in 3D space that is modeled using a cloud-specific neural network. To capture the similarity between point clouds we rely on parameter sharing among networks, with each cloud having only a small embedding vector that defines it. We use invertible flows networks to generate the individual point clouds, and to regularize the embedding vectors. We evaluate the generative capabilities of the model both in qualitative and quantitative manner.

cs.LG↗

arXiv2019

Lingvo: a Modular and Scalable Framework for Sequence-to-Sequence Modeling

Lingvo is a Tensorflow framework offering a complete solution for collaborative deep learning research, with a particular focus towards sequence-to-sequence models. Lingvo models are composed of modular building blocks that are flexible and easily extensible, and experiment configurations are centralized and highly customizable. Distributed training and quantized inference are supported directly within the framework, and it contains existing implementations of a large number of utilities, helper functions, and the newest research ideas. Lingvo has been used in collaboration by dozens of researchers in more than 20 papers over the last two years. This document outlines the underlying design of Lingvo and serves as an introduction to the various pieces of the framework, while also offering examples of advanced features that showcase the capabilities of the framework.

cs.LG↗

arXiv2019

Unsupervised speech representation learning using WaveNet autoencoders

We consider the task of unsupervised extraction of meaningful latent representations of speech by applying autoencoding neural networks to speech waveforms. The goal is to learn a representation able to capture high level semantic content from the signal, e.g.\ phoneme identities, while being invariant to confounding low level details in the signal such as the underlying pitch contour or background noise. Since the learned representation is tuned to contain only phonetic content, we resort to using a high capacity WaveNet decoder to infer information discarded by the encoder from previous samples. Moreover, the behavior of autoencoder models depends on the kind of constraint that is applied to the latent representation. We compare three variants: a simple dimensionality reduction bottleneck, a Gaussian Variational Autoencoder (VAE), and a discrete Vector Quantized VAE (VQ-VAE). We analyze the quality of learned representations in terms of speaker independence, the ability to predict phonetic content, and the ability to accurately reconstruct individual spectrogram frames. Moreover, for discrete encodings extracted using the VQ-VAE, we measure the ease of mapping them to phonemes. We introduce a regularization scheme that forces the representations to focus on the phonetic content of the utterance and report performance comparable with the top entries in the ZeroSpeech 2017 unsupervised acoustic unit discovery task.

cs.LG↗

arXiv2019

Towards Using Context-Dependent Symbols in CTC Without State-Tying Decision Trees

Deep neural acoustic models benefit from context-dependent (CD) modeling of output symbols. We consider direct training of CTC networks with CD outputs, and identify two issues. The first one is frame-level normalization of probabilities in CTC, which induces strong language modeling behavior that leads to overfitting and interference with external language models. The second one is poor generalization in the presence of numerous lexical units like triphones or tri-chars. We mitigate the former with utterance-level normalization of probabilities. The latter typically requires reducing the CD symbol inventory with state-tying decision trees, which have to be transferred from classical GMM-HMM systems. We replace the trees with a CD symbol embedding network, which saves parameters and ensures generalization to unseen and undersampled CD symbols. The embedding network is trained together with the rest of the acoustic model and removes one of the last cases in which neural systems have to be bootstrapped from GMM-HMM ones.

cs.CL↗

arXiv2018

Efficient Purely Convolutional Text Encoding

In this work, we focus on a lightweight convolutional architecture that creates fixed-size vector embeddings of sentences. Such representations are useful for building NLP systems, including conversational agents. Our work derives from a recently proposed recursive convolutional architecture for auto-encoding text paragraphs at byte level. We propose alternations that significantly reduce training time, the number of parameters, and improve auto-encoding accuracy. Finally, we evaluate the representations created by our model on tasks from SentEval benchmark suite, and show that it can serve as a better, yet fairly low-resource alternative to popular bag-of-words embeddings.

cs.CL↗

arXiv2018

A Talker Ensemble: the University of Wrocław's Entry to the NIPS 2017 Conversational Intelligence Challenge

We present Poetwannabe, a chatbot submitted by the University of Wrocław to the NIPS 2017 Conversational Intelligence Challenge, in which it ranked first ex-aequo. It is able to conduct a conversation with a user in a natural language. The primary functionality of our dialogue system is context-aware question answering (QA), while its secondary function is maintaining user engagement. The chatbot is composed of a number of sub-modules, which independently prepare replies to user's prompts and assess their own confidence. To answer questions, our dialogue system relies heavily on factual data, sourced mostly from Wikipedia and DBpedia, data of real user interactions in public forums, as well as data concerning general literature. Where applicable, modules are trained on large datasets using GPUs. However, to comply with the competition's requirements, the final system is compact and runs on commodity hardware.

cs.CL↗

arXiv2018

On Using Backpropagation for Speech Texture Generation and Voice Conversion

Inspired by recent work on neural network image generation which rely on backpropagation towards the network inputs, we present a proof-of-concept system for speech texture synthesis and voice conversion based on two mechanisms: approximate inversion of the representation learned by a speech recognition neural network, and on matching statistics of neuron activations between different source and target utterances. Similar to image texture synthesis and neural style transfer, the system works by optimizing a cost function with respect to the input waveform samples. To this end we use a differentiable mel-filterbank feature extraction pipeline and train a convolutional CTC speech recognition network. Our system is able to extract speaker characteristics from very limited amounts of target speaker data, as little as a few seconds, and can be used to generate realistic speech babble or reconstruct an utterance in a different voice.

cs.SD↗