Analogical Reasoning Inside Large Language Models: Concept Vectors and the Limits of Abstraction

Claire E. Stevenson; Gustaw Opie{\l}ka; Hannes Rosenbusch

Analogical Reasoning Inside Large Language Models: Concept Vectors and the Limits of Abstraction

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2503.03666 v1 pith:WGNRM7V5 submitted 2025-03-05 cs.CL cs.LG

Gustaw Opie{l}ka , Hannes Rosenbusch , Claire E. Stevenson This is my paper

classification cs.CLcs.LG

keywords representationsconceptsinvariantvectorsanalogicalconceptfunctioninternal

verification ladder T0 review T1 audit T2 compute T3 formal

0 comments

read the original abstract

Analogical reasoning relies on conceptual abstractions, but it is unclear whether Large Language Models (LLMs) harbor such internal representations. We explore distilled representations from LLM activations and find that function vectors (FVs; Todd et al., 2024) - compact representations for in-context learning (ICL) tasks - are not invariant to simple input changes (e.g., open-ended vs. multiple-choice), suggesting they capture more than pure concepts. Using representational similarity analysis (RSA), we localize a small set of attention heads that encode invariant concept vectors (CVs) for verbal concepts like "antonym". These CVs function as feature detectors that operate independently of the final output - meaning that a model may form a correct internal representation yet still produce an incorrect output. Furthermore, CVs can be used to causally guide model behaviour. However, for more abstract concepts like "previous" and "next", we do not observe invariant linear representations, a finding we link to generalizability issues LLMs display within these domains.

Discussion (0). Sign in to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

Metaphor-Induced Algorithmic Steering: Cross-Domain Procedural Transfer in LLM Code Generation
cs.SE 2026-07 conditional novelty 7.0 of 10

Metaphorical instructions can steer LLM code generators toward correct but less efficient algorithms, and this steering is reflected in hidden-state representations.
Analogical Deep Research: Retrieving and Integrating Historical Analogies for Foresight Analysis
cs.CL 2026-07 conditional novelty 6.0 of 10

LLM deep-research agents rarely use historical analogies; a structural-decomposition plus cross-analogy-confirmation agent (CANA) sharply increases mechanism-grounded analogy claims and hidden-factor hits on the new A...
Emergent Structured Representations Support Flexible In-Context Inference in Large Language Models
cs.CL 2026-02 unverdicted novelty 6.0 of 10

LLMs dynamically construct and causally rely on structured conceptual subspaces in middle-to-late layers for in-context inference.
Parallel LLM Reasoning for Bias-Resilient, Robust Conceptual Abstraction
cs.CL 2026-04 unverdicted novelty 4.0 of 10

Parallel chunk processing with evidence-anchored consolidation reduces omission errors by 84%, boosts traceability by 130%, and cuts unsupported claims by 91% in LLM long-document analysis.

Pith tools