arXiv AI September 28, 2026 By Karthik Singaravadivelan, Pat Langley

A Unified Account of Concepts and Chunks

The paper reviews Cobweb, a computational model of categorization and concept formation, and extends it to include chunks and their acquisition. It introduces rellis/, an implementation that applies this unified theory to learning context-free grammars, demonstrating the system’s ability to represent syntactic knowledge, parse and generate sentences, and learn compositional structures from sample parses. The authors discuss related work on concepts and chunks and suggest directions for future research.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv Machine Learning

Jun 11

Learning Patterns and Abstractions from Perceptual Sequences

arXiv:2503. 10973v2 Announce Type: replace Abstract: Cognition swiftly breaks high-dimensional sensory streams into familiar parts and uncovers their relations.

By Shuchen Wu

llms rag

arXiv AI

3d ago

Functional Subspace, where language models can use vector algebra to solve problems

The paper proposes that large language models (LLMs) encode high‑level concepts as linear directions within their activation space and that they can use subspaces and vector algebra to perform tasks. By analyzing functional modules and residual streams during in‑context learning (ICL), the authors find that LLMs can create evidence‑accumulating subspaces and solve ICL tasks through simple algebraic operations within those subspaces.

By Jung H. Lee, Sujith Vijayan

llms rag

arXiv AI

Jul 22

From Dependency to Compositionality: A Neurosymbolic Lifting of LLM Outputs via Combinatory Categorial Grammar

arXiv:2607. 18961v1 Announce Type: new Abstract: Large language models (LLMs) generate fluent text by incrementally predicting the next token from a prefix.

By Remo Pareschi

llms data-engineering

arXiv AI

Sep 7

Towards a universal language of concepts: A survey

The paper surveys computational models that use programs as representations for concepts, arguing that programs could serve as a universal language for concepts. It highlights how humans learn and generalize from sparse data by expressing knowledge in rich structural formats, and evaluates how program-based models contribute to this goal. The authors propose that adopting programs as a universal representational language could enhance concept learning across diverse domains.

By Aishni Parab

arXiv AI

Jul 28

The Lattice Representation Hypothesis of Large Language Models

arXiv:2603. 01227v3 Announce Type: replace Abstract: We propose the Lattice Representation Hypothesis of large language models: a symbolic backbone that grounds conceptual hierarchies and logical operations in embedding geometry.

By Bo Xiong

llms rag

arXiv Machine Learning

Jun 25

Weave of Formal Thought

arXiv:2606. 25987v1 Announce Type: cross Abstract: Large language models (LLMs) attain remarkable surface fluency on code, yet they neither formally guarantee the syntactic validity of their output nor leverage the hierarchical structure defining the target language.

By Alexandre Bouayad

llms nlp fine-tuning

One email a morning, machine-written

Related stories

Learning Patterns and Abstractions from Perceptual Sequences

Functional Subspace, where language models can use vector algebra to solve problems

From Dependency to Compositionality: A Neurosymbolic Lifting of LLM Outputs via Combinatory Categorial Grammar

Towards a universal language of concepts: A survey

The Lattice Representation Hypothesis of Large Language Models

Weave of Formal Thought