arXiv research

Unified causal model improves controllable text generation without bias.

Improved text generation with constraints using discrete auto-regressive biasing.

New text-to-image diffusion models improve scene understanding for AI agents.

Generic generation and manipulation of text is challenging and has limited success compared to recent deep generative modeling in visual domain. This paper aims at generating plausible natural language sentences, whose attributes are dynamically controlled by learning disentangled latent representations with designated…

Unified framework for generating meteorological time series from text.

Data availability is a bottleneck during early stages of development of new capabilities for intelligent artificial agents. We investigate the use of text generation techniques to augment the training data of a popular commercial artificial agent across categories of functionality, with the goal of faster development o…

Improved text-to-image and multimodal understanding through adaptive generation order optimization.

Improved image synthesis with user scribbles and text prompts.

One important partition of algorithms for controlling the false discovery rate (FDR) in multiple testing is into offline and online algorithms. The first generally achieve significantly higher power of discovery, while the latter allow making decisions sequentially as well as adaptively formulating hypotheses based on …

Paper develops heavy-tailed embeddings for better text classification and augmentation.

We introduce TextWorld, a sandbox learning environment for the training and evaluation of RL agents on text-based games. TextWorld is a Python library that handles interactive play-through of text games, as well as backend functions like state tracking and reward assignment. It comes with a curated list of games whose …

Method detects if text is generated by a language model with watermarks.

Conditional Text Generation has drawn much attention as a topic of Natural Language Generation (NLG) which provides the possibility for humans to control the properties of generated contents. Current conditional generation models cannot handle emerging conditions due to their joint end-to-end learning fashion. When a n…

Grad-TTS models speech from text using diffusion probabilistic techniques.

Generative autoencoders offer a promising approach for controllable text generation by leveraging their latent sentence representations. However, current models struggle to maintain coherent latent spaces required to perform meaningful text manipulations via latent vector operations. Specifically, we demonstrate by exa…

Classifier detects LLM-generated text with guarantees.

Deep generative models are commonly used for generating images and text. Interpretability of these models is one important pursuit, other than the generation quality. Variational auto-encoder (VAE) with Gaussian distribution as prior has been successfully applied in text generation, but it is hard to interpret the mean…

New framework improves text watermark detection under imperfect pseudorandomness.

Study resolvent convergence for random matrices with general covariance profiles.

Efficiently controls unknown linear systems with black-box interactions.

Improves text-to-image diffusion models using GFlowNets.

Self-guidance controls image generation by extracting properties from diffusion model representations.

Reinforcement learning algorithms, though successful, tend to over-fit to training environments hampering their application to the real-world. This paper proposes $\text{W}\text{R}^{2}\text{L}$ -- a robust reinforcement learning algorithm with significant robust performance on low and high-dimensional control tasks. Ou…

In this work we propose a new computational framework, based on generative deep models, for synthesis of photo-realistic food meal images from textual descriptions of its ingredients. Previous works on synthesis of images from text typically rely on pre-trained text models to extract text features, followed by a genera…

AdaptOn achieves logarithmic regret in adaptive control of unknown partially observable linear systems.

ControlVAE improves VAE performance by adding a controller to tune hyperparameters.

We study the problem of controllable generation of long-term sequential behaviors, where the goal is to calibrate to multiple behavior styles simultaneously. In contrast to the well-studied areas of controllable generation of images, text, and speech, there are two questions that pose significant challenges when genera…

The paper formalizes how concepts are encoded in text-guided generative models and provides a method to manipulate them.

Paper introduces a statistical framework for watermarking LLM-generated text.

Automatic voice-controlled systems have changed the way humans interact with a computer. Voice or speech recognition systems allow a user to make a hands-free request to the computer, which in turn processes the request and serves the user with appropriate responses. After years of research and developments in machine …

RB-Modulation trains free diffusion models without external adapters.

PAC-Bayesian theory improves text-to-image models by enforcing alignment and generalization.

Embed-KCPD segments text without labels, outperforming baselines.

In this work, we address the problem of modifying textual attributes of sentences. Given an input sentence and a set of attribute labels, we attempt to generate sentences that are compatible with the conditioning information. To ensure that the model generates content compatible sentences, we introduce a reconstruction…

HRT uses bi-level reinforcement learning to optimize stock selection and execution in multi-asset equity markets.

We present a syntax-infused variational autoencoder (SIVAE), that integrates sentences with their syntactic trees to improve the grammar of generated sentences. Distinct from existing VAE-based text generative models, SIVAE contains two separate latent spaces, for sentences and syntactic trees. The evidence lower bound…

Unified model for audio control and style transfer.

Sparse text alignments learned via optimal transport improve model explainability.

Generating versatile and appropriate synthetic speech requires control over the output expression separate from the spoken text. Important non-textual speech variation is seldom annotated, in which case output control must be learned in an unsupervised fashion. In this paper, we perform an in-depth study of methods for…

ICP improves text infilling and POS tagging with valid confidence sets.

Discrete diffusion models improve text and image inference.

Paper introduces Latent-CLIP for efficient text-image comparison in latent space.

Framework for ensuring fairness in machine learning models across multiple groups.

Optimal algorithm for LQR control with improved regret bound.

Scarcity of training data for task-oriented dialogue systems is a well known problem that is usually tackled with costly and time-consuming manual data annotation. An alternative solution is to rely on automatic text generation which, although less accurate than human supervision, has the advantage of being cheap and f…

LIMEADE improves AI advice for opaque models, enhancing accuracy and user satisfaction.

New control methods for systems with adversarial perturbations.

While perception tasks such as visual object recognition and text understanding play an important role in human intelligence, the subsequent tasks that involve inference, reasoning and planning require an even higher level of intelligence. The past few years have seen major advances in many perception tasks using deep …