Research
On-device research index

arXiv research

A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.

168,695 papers · 148 categories

Trend · papers per month

72143215286 · Jun 202019922001200920172026
48 results for discrete attention

A new model tackles language generation issues by using discrete variational attention.

problem Information under-representation and posterior collapse in variational autoencoders.
method Proposes a discrete variational attention model with categorical distribution over attention mechanism.
result Enhances latent space for language generation and avoids posterior collapse.

A new model captures complex event data using attention and Fourier kernels.

problem Capturing complex non-linear temporal dependencies in discrete event data.
method Integrates attention mechanism into point processes' conditional intensity function and uses Fourier kernel embedding.
result Established theoretical properties and demonstrated competitive performance.

Sequence-to-sequence text-to-speech (TTS) is dominated by soft-attention-based methods. Recently, hard-attention-based methods have been proposed to prevent fatal alignment errors, but their sampling method of discrete alignment is poorly investigated. This research investigates various combinations of sampling methods…

2019-10-28abs ↗pdf ↗

SympFormer accelerates attention blocks using inertial dynamics on density spaces.

problem Improving the efficiency of self-attention blocks in Transformers.
method Introduced accelerated attention blocks derived from inertial Nesterov dynamics on density spaces.
result Accelerated attention blocks converge faster than classical blocks while preserving oracle calls.

New approach improves linear-time attention for language models.

problem Challenges of quadratic attention in long-sequence modelling, especially for discrete data.
method Reinterpreting linear attention through latent probabilistic graphical models, introducing asymmetric structure and recurrent parameterisation.
result Our model achieves competitive performance and outperforms existing linear attention variants on language modelling benchmarks.

Paper proposes Sinkformers for Transformers with doubly stochastic attention.

problem Improving Transformer models' accuracy in vision and natural language processing.
method Using Sinkhorn's algorithm to make attention matrices doubly stochastic instead of SoftMax normalization.
result Sinkformers enhance model accuracy in vision and natural language processing tasks.

PDHAMS improves sampling for discrete distributions with quadratic potential functions.

problem Sampling discrete distributions efficiently and accurately.
method Integrates a second-order approximation of the potential function and uses Gaussian integral trick.
result PDHAMS yields superior performance compared to other methods.

We introduce a new neural architecture to learn the conditional probability of an output sequence with elements that are discrete tokens corresponding to positions in an input sequence. Such problems cannot be trivially addressed by existent approaches such as sequence-to-sequence and Neural Turing Machines, because th…

2015-06-09abs ↗pdf ↗

There has recently been significant interest in hard attention models for tasks such as object recognition, visual captioning and speech recognition. Hard attention can offer benefits over soft attention such as decreased computational cost, but training hard attention models can be difficult because of the discrete la…

2017-05-16abs ↗pdf ↗

Unified discrete diffusion for categorical data simplifies training and sampling.

problem Training and sampling in discrete diffusion models for categorical data.
method Mathematical simplifications and elegant unification of discrete-time and continuous-time discrete diffusion.
result Unified Simplified Discrete Denoising Diffusion (USD3) outperforms SOTA baselines.

Paper extends sparse alternatives to softmax for continuous domains, enabling efficient attention mechanisms.

problem Efficiently assigning zero probability to irrelevant categories in continuous domains.
method Extend alpha-entmax to continuous domains, introducing continuous-domain attention mechanisms.
result Continuous attention allows attending to time intervals and compact regions, improving text classification, machine translation, and visual question answering.

This work analyzes discrete diffusion models using stochastic integrals, providing error bounds and insights.

problem Error analysis for discrete diffusion models remains less understood.
method Proposes a comprehensive framework based on Lévy-type stochastic integrals.
result Obtains the first error bound for the ττ-leaping scheme in KL divergence.

Generalized meshes for non-regular geometries, including fractures.

problem Discretization of partial differential equations in non-regular geometries.
method Introduces generalized meshes with overlapping elements and flexible adjacency relations.
result Discrete differential forms on virtually inflated meshes characterize the trace space of forms in surrounding volumes.

A study on optimizing self-attention in tabular data using Optimal Transport.

problem Improving efficiency and accuracy of self-attention in tabular classification tasks.
method Developed an OT-based algorithm to generate class-specific dummy Gaussian distributions and train an MLP.
result Achieved comparable accuracy to Transformers with reduced computational cost and efficiency.

Transformers cluster meaningless words around leaders for sentiment analysis.

problem Capturing context in sentiment analysis using transformers.
method Characterized transformers with hardmax self-attention and normalization, showing asymptotic convergence to clustered equilibrium.
result Transformers can effectively capture context by clustering meaningless words around leader words.

In this article we study the role of the Green function for the Laplacian in a compact Riemannian manifold as a tool for obtaining well-distributed points. In particular, we prove that a sequence of minimizers for the Green energy is asymptotically uniformly distributed. We pay special attention to the case of locally …

2017-02-02abs ↗pdf ↗

GLAD improves latent graph generation by quantizing discrete latent space.

problem Latent space graph generative models lack performance and make unnatural assumptions.
method Adapting diffusion bridges to a discrete latent space, avoiding data space decompositions.
result GLAD achieves competitive performance on graph benchmark datasets.

Unified continuous diffusion model outperforms discrete alternatives in scalability and quality.

problem Continuous diffusion models were perceived as less scalable than discrete models.
method Reconstructed Plaid model and compared it with modern discrete DLMs, optimizing noise schedule and embeddings via likelihood.
result Unified continuous diffusion model (RePlaid) outperforms discrete models in compute efficiency and quality.

Study on deep multi-head self-attention dynamics, proving homogenized limits under specific scalings.

problem Understanding the behavior of deep multi-head self-attention models as depth increases.
method Random model of deep multi-head self-attention, viewing depth as time, and analyzing the residual stream as a particle system.
result Homogenized limit of the dynamics, leading to deterministic or stochastic behavior depending on scaling, with implications for representation collapse.

Bayesian optimization tackles mixed discrete-continuous problems with Gaussian processes.

problem Optimizing problems with both discrete and continuous variables using costly simulations.
method Relaxing discrete variables into continuous latent variables, using Bayesian optimization, and incorporating compatibility constraints with Lagrangians.
result Comparative analysis of different mixed Bayesian optimization approaches.

Stochastic optimization improves semi-discrete OT map estimation with a minimax rate.

problem Empirical success of SGD in semi-discrete OT, but lack of theoretical guarantees.
method Averaged projected SGD with a minimax convergence rate of O(1/√n).
result SGD methods can estimate the OT map with a minimax convergence rate of O(1/√n).

A new optimizer preserves orthogonality constraints on matrices efficiently.

problem Optimization on Stiefel manifold with orthogonality constraints.
method Interplay between continuous and discrete dynamics leading to a gradient-based optimizer with momentum.
result The method optimizes matrices on Stiefel manifold efficiently and accurately.

Uniform scaling limits in AdamW-trained transformers converge to ODEs.

problem Understanding the dynamics of large-depth transformers trained with AdamW.
method Modeling transformer dynamics as an interacting particle system coupled through attention, proving convergence to ODEs.
result The joint dynamics of hidden states and backpropagated variables converge uniformly to an ODE system.

Numerous methods for crafting adversarial examples were proposed recently with high success rate. Since most existing machine learning based classifiers normalize images into some continuous, real vector, domain firstly, attacks often craft adversarial examples in such domain. However, "adversarial" examples may become…

2019-05-19abs ↗pdf ↗

Data-dependent hashing has recently attracted attention due to being able to support efficient retrieval and storage of high-dimensional data such as documents, images, and videos. In this paper, we propose a novel learning-based hashing method called "Supervised Discrete Hashing with Relaxation" (SDHR) based on "Super…

2019-04-07abs ↗pdf ↗

Proposes a method to quantify uncertainty in DNN models for discrete inputs.

problem Uncertainty quantification for DNN models with categorical and discrete feature variables.
method Develops a mathematical framework to quantify prediction uncertainty from discrete input noise and model parameters.
result Identifies risk-sensitive cases prone to misclassification due to discrete predictor errors.

LEAPS samples discrete distributions via CTMCs and locally equivariant networks.

problem Sampling from discrete distributions with known normalization.
method Continuous-time Markov chain, locally equivariant functions, attention layers, convolutional networks.
result LEAPS minimizes the variance of importance weights, improving sampling efficiency.

This work studies learning a multi-head attention layer from random examples.

problem Learning a multi-head attention layer from random examples.
method The work initiates the study of provably learning a multi-head attention layer from random examples, providing upper and lower bounds.
result The first nontrivial upper and lower bounds for learning a multi-head attention layer from random examples are given.

Recurrent Neural Networks (RNNs) with attention mechanisms have obtained state-of-the-art results for many sequence processing tasks. Most of these models use a simple form of encoder with attention that looks over the entire sequence and assigns a weight to each token independently. We present a mechanism for focusing…

2018-06-12abs ↗pdf ↗

Transformers can predict new tokens based on any number of context tokens, approximating continuous mappings with fixed resources.

problem Handling an arbitrarily large number of context tokens in transformers.
method Mathematical analysis of transformer's expressivity using Wasserstein distance and continuous mappings.
result Deep transformers are universal and can approximate continuous in-context mappings to arbitrary precision, uniformly over compact token domains.

Many transformations in deep learning architectures are sparsely connected. When such transformations cannot be designed by hand, they can be learned, even through plain backpropagation, for instance in attention mechanisms. However, during learning, such sparse structures are often represented in a dense form, as we d…

2018-10-22abs ↗pdf ↗