Parallelograms Strike Back: LLMs Generate Better Analogies than People

• URL: http://arxiv.org/abs/2603.19066v1
• Date: Thu, 19 Mar 2026 15:56:24 GMT
• Title: Parallelograms Strike Back: LLMs Generate Better Analogies than People
• Abstract summary: We compare human and large language model (LLM) analogy completions on analogy problems.<n>We find that LLM-generated analogies are reliably judged as better than human-generated ones.
• Score: 9.954471631773353
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Four-term word analogies (A:B::C:D) are classically modeled geometrically as ''parallelograms,'' yet recent work suggests this model poorly captures how humans produce analogies, with simple local-similarity heuristics often providing a better account (Peterson et al., 2020). But does the parallelogram model fail because it is a bad model of analogical relations, or because people are not very good at generating relation-preserving analogies? We compared human and large language model (LLM) analogy completions on the same set of analogy problems from (Peterson et al., 2020). We find that LLM-generated analogies are reliably judged as better than human-generated ones, and are also more closely aligned with the parallelogram structure in a distributional embedding space (GloVe). Crucially, we show that the improvement over human analogies was driven by greater parallelogram alignment and reduced reliance on accessible words rather than enhanced sensitivity to local similarity. Moreover, the LLM advantage is driven not by uniformly superior responses by LLMs, but by humans producing a long tail of weak completions: when only modal (most frequent) responses by both systems are compared, the LLM advantage disappears. However, greater parallelogram alignment and lower word frequency continue to predict which LLM completions are rated higher than those of humans. Overall, these results suggest that the parallelogram model is not a poor account of word analogy. Rather, humans may often fail to produce completions that satisfy this relational constraint, whereas LLMs do so more consistently.

Related papers

• Will Scaling Improve Social Simulation with LLMs? [87.15677770205751]
  Large Language Model (LLM) social simulations are a promising research method, but they are not yet faithful enough to be adopted widely.<n>We use scaling laws to study the relationship between LLMs' compute scale, general capability benchmarks, and the fidelity of social simulation.
  arXiv  Detail & Related papers  (2026-07-02T17:30:38Z)
• Demystifying Pipeline Parallelism: First Theory for PipeDream [53.657104889705856]
  In this paper we introduce Randomized PipeDream (PD) as a stale block-SGD abstraction that yields clean non convergence for a PD-style method.<n>We prove that the delay induced by steady-state PD grows as $S2 - S/2 + O(1)$ for $S$ stages, so the stale-read contribution scales as $(2 S4)$, equivalently $(S4/K)$ in the tuned-rate form.
  arXiv  Detail & Related papers  (2026-06-02T11:14:57Z)
• Towards Real-world Human Behavior Simulation: Benchmarking Large Language Models on Long-horizon, Cross-scenario, Heterogeneous Behavior Traces [81.41397370235102]
  We introduce OmniBehavior, the first user simulation benchmark constructed entirely from real-world data.<n>We show that current models struggle to accurately simulate complex behaviors, with performance plateauing even as context windows expand.<n>This results in the loss of individual differences and long-tail behaviors, highlighting critical directions for future high-fidelity simulation research.
  arXiv  Detail & Related papers  (2026-04-09T15:26:21Z)
• Measuring Scalar Constructs in Social Science with LLMs [48.92998035333579]
  We evaluate approaches to measuring scalar constructs in large language models.<n>We find that pairwise comparisons produce better measurements than simply prompting the LLM to directly output the scores.<n>Finetuning smaller models with as few as 1,000 training pairs can match or exceed the performance of prompted LLMs.
  arXiv  Detail & Related papers  (2025-09-03T08:19:13Z)
• Modeling Understanding of Story-Based Analogies Using Large Language Models [1.4999444543328293]
  Recent advancements in Large Language Models have brought them closer to matching human cognition across a variety of tasks.<n>How well do these models align with human performance in detecting and mapping analogies?
  arXiv  Detail & Related papers  (2025-07-15T03:40:21Z)
• Can LLMs Help Improve Analogical Reasoning For Strategic Decisions? Experimental Evidence from Humans and GPT-4 [0.0]
  This study investigates whether large language models, specifically GPT4, can match human capabilities in analogical reasoning.<n>Using a novel experimental design involving source to target matching, we find that GPT4 achieves high recall by retrieving all plausible analogies but suffers from low precision.<n>Human participants exhibit high precision but low recall, selecting fewer analogies yet with stronger causal alignment.
  arXiv  Detail & Related papers  (2025-05-01T15:35:01Z)
• QUDsim: Quantifying Discourse Similarities in LLM-Generated Text [70.22275200293964]
  We introduce an abstraction based on linguistic theories in Questions Under Discussion (QUD) and question semantics to help quantify differences in discourse progression.<n>We then use this framework to build $textbfQUDsim$, a similarity metric that can detect discursive parallels between documents.<n>Using QUDsim, we find that LLMs often reuse discourse structures (more so than humans) across samples, even when content differs.
  arXiv  Detail & Related papers  (2025-04-12T23:46:09Z)
• Is analogy enough to draw novel adjective-noun inferences? [9.30694340695458]
  We study whether inferences can instead be derived by analogy to known inferences, without need for composition.<n>We find that there are novel combinations for which both humans and LLMs derive convergent inferences but which are not well handled by analogy.
  arXiv  Detail & Related papers  (2025-03-31T16:41:16Z)
• Do Large Language Models Solve ARC Visual Analogies Like People Do? [0.0]
  We compared human and large language model (LLM) performance on a new child-friendly set of ARC items. Results show that both children and adults outperform most LLMs on these tasks.
  arXiv  Detail & Related papers  (2024-03-13T09:48:13Z)
• ParallelPARC: A Scalable Pipeline for Generating Natural-Language Analogies [16.92480305308536]
  We develop a pipeline for creating complex, paragraph-based analogies. We publish a gold-set, validated by humans, and a silver-set, generated automatically. We demonstrate that our silver-set is useful for training models.
  arXiv  Detail & Related papers  (2024-03-02T08:53:40Z)
• StoryAnalogy: Deriving Story-level Analogies from Large Language Models to Unlock Analogical Understanding [72.38872974837462]
  We evaluate the ability to identify and generate analogies by constructing a first-of-its-kind large-scale story-level analogy corpus. textscStory Analogy contains 24K story pairs from diverse domains with human annotations on two similarities from the extended Structure-Mapping Theory. We observe that the data in textscStory Analogy can improve the quality of analogy generation in large language models.
  arXiv  Detail & Related papers  (2023-10-19T16:29:23Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.