Article

Zero-shot learning of naturalistic concepts via conceptual systems alignment

2026-06-26

Abstract excerpt

<p>Words that are used similarly tend to point to referents that share perceptual features. Because humans are sensitive to these second-order similarities or systems alignments, they might be able to leverage representations available in one system (how a concept is talked about) to infer those lacking in another (how it might look like). Testing this form of zero-shot learning in humans is, however, tricky: conc...

Topics

Open a Topic to create a Post that cites this publication.

Identifiers and source

Literature Corpus work: 3a53b464-c042-52e5-8623-9abc9660bb17
DOI: 10.31234/osf.io/4e8un_v1

Open publication

Related research

Semantic proximity does not establish scientific evidence.

Click a neighbor to travelStep 1 · 12 closest

Zero-shot learning of naturalistic concepts via conceptual systems alignmentDOI 10.31234/osf.io/4e8un_v1Read

Towards Cognition-Aligned Visual Language Models via Zero-Shot Instance RetrievalSemantic proximity Read

System alignment supports cross-domain learning and zero-shot generalisationSemantic proximity Read

Leveraging Context for Perceptual Prediction using Word EmbeddingsSemantic proximity Read

Using Compositionality to Learn Many Categories from Few ExamplesSemantic proximity Read

Between semasiology and onomasiology: leveraging meaning-form mapping to create novel words with a predefined meaning.Semantic proximity Read

Cross-linguistic zero-shot communication via ad-hoc pseudowordsSemantic proximity Read

Meaning Beyond Lexicality: Capturing Pseudoword Definitions with Language ModelsSemantic proximity Read

Zero-shot pseudowords memorability via representational content analysisSemantic proximity Read

Form Composition: The Generative Structural Engine of Metaphors in Human CognitionSemantic proximity Read