Related papers: Vector Generation of Contextual Sets

Generating inferential texts about an event in different perspectives requires reasoning over different contexts that the event occurs. Existing works usually ignore the context that is not explicitly provided, resulting in a…

There has been exciting progress in generating images from natural language or layout conditions. However, these methods struggle to faithfully reproduce complex scenes due to the insufficient modeling of multiple objects and their…

Previous works on knowledge-to-text generation take as input a few RDF triples or key-value pairs conveying the knowledge of some entities to generate a natural language description. Existing datasets, such as WIKIBIO, WebNLG, and E2E,…

In the paper it is shown that the Kochen-Specker theorem follows from Burnside's theorem on noncommutative algebras. Accordingly, contextuality (as an impossibility of assigning binary values to projection operators independently of their…

An important component of achieving language understanding is mastering the composition of sentence meaning, but an immediate challenge to solving this problem is the opacity of sentence vector representations produced by current neural…

We present a neural model for generating short stories from image sequences, which extends the image description model by Vinyals et al. (Vinyals et al., 2015). This extension relies on an encoder LSTM to compute a context vector of each…

This work introduces an enhanced approach to generating scene graphs by incorporating both a relationship hierarchy and commonsense knowledge. Specifically, we begin by proposing a hierarchical relation head that exploits an informative…

Kernel methods are a highly effective and widely used collection of modern machine learning algorithms. A fundamental limitation of virtually all such methods are computations involving the kernel matrix that naively scale quadratically…

Background / introduction. Vector symbolic architectures (VSA) are a viable approach for the hyperdimensional representation of symbolic data, such as documents, syntactic structures, or semantic frames. Methods. We present a rigorous…

Kochen-Specker theorem rules out the non-contextual assignment of values to physical magnitudes. Here we enrich the usual orthomodular structure of quantum mechanical propositions with modal operators. This enlargement allows to refer…

We consider algebras underlying Hilbert spaces used by quantum information algorithms. We show how one can arrive at equations on such algebras which define n-dimensional Hilbert space subspaces which in turn can simulate quantum systems on…

We present a novel natural language generation system for spoken dialogue systems capable of entraining (adapting) to users' way of speaking, providing contextually appropriate responses. The generator is based on recurrent neural networks…

The question of a hidden variable interpretation of quantum contextuality in the Mermin-Peres square is considered. The Kochen-Specker theorem implies that quantum mechanics may be interpreted as a contextual hidden variable theory. It is…

Generating realistic images from scene graphs asks neural networks to be able to reason about object relationships and compositionality. As a relatively new task, how to properly ensure the generated images comply with scene graphs or how…

We propose an experimental approach to {\it macro}scopically test the Kochen-Specker theorem (KST) with superconducting qubits. This theorem, which has been experimentally tested with single photons or neutrons, concerns the conflict…

We demonstrate that pre-trained text-to-image diffusion models, despite being trained on raster images, possess a remarkable capacity to guide vector sketch synthesis. In this paper, we introduce DiffSketcher, a novel algorithm for…

Quantum systems show contextuality. More precisely, it is impossible to reproduce the quantum-mechanical predictions using a non-contextual realist model, i.e., a model where the outcome of one measurement is independent of the choice of…

We present a novel response generation system that can be trained end to end on large quantities of unstructured Twitter conversations. A neural network architecture is used to address sparsity issues that arise when integrating contextual…

The integration of knowledge graphs (KGs) with large language models (LLMs) offers significant potential to improve the retrieval phase of retrieval-augmented generation (RAG) systems. In this study, we propose KG-CQR, a novel framework for…

The dominant text generation models compose the output by sequentially selecting words from a fixed vocabulary. In this paper, we formulate text generation as progressively copying text segments (e.g., words or phrases) from an existing…