SP
BravenNow
Enhancing Structural Mapping with LLM-derived Abstractions for Analogical Reasoning in Narratives
| USA | technology | ✓ Verified - arxiv.org

Enhancing Structural Mapping with LLM-derived Abstractions for Analogical Reasoning in Narratives

📖 Full Retelling

arXiv:2603.29997v1 Announce Type: cross Abstract: Analogical reasoning is a key driver of human generalization in problem-solving and argumentation. Yet, analogies between narrative structures remain challenging for machines. Cognitive engines for structural mapping are not directly applicable, as they assume pre-extracted entities, whereas LLMs' performance is sensitive to prompt format and the degree of surface similarity between narratives. This gap motivates a key question: What is the impa

📚 Related People & Topics

Artificial intelligence

Artificial intelligence

Intelligence of machines

# Artificial Intelligence (AI) **Artificial Intelligence (AI)** is a specialized field of computer science dedicated to the development and study of computational systems capable of performing tasks typically associated with human intelligence. These tasks include learning, reasoning, problem-solvi...

View Profile → Wikipedia ↗

Large language model

Type of machine learning model

A large language model (LLM) is a language model trained with self-supervised machine learning on a vast amount of text, designed for natural language processing tasks, especially language generation. The largest and most capable LLMs are generative pre-trained transformers (GPTs) that provide the c...

View Profile → Wikipedia ↗

Entity Intersection Graph

Connections for Artificial intelligence:

🏢 OpenAI 14 shared
🌐 Reinforcement learning 4 shared
🏢 Anthropic 4 shared
🌐 Large language model 3 shared
🏢 Nvidia 3 shared
View full profile

Mentioned Entities

Artificial intelligence

Artificial intelligence

Intelligence of machines

Large language model

Type of machine learning model

Deep Analysis

Why It Matters

This research matters because it advances how artificial intelligence can understand and reason with complex narratives, which has implications for education, content analysis, and decision support systems. It affects educators who could use such systems to teach analytical thinking, researchers in computational linguistics and cognitive science, and developers building AI tools for legal analysis, literary studies, or strategic planning. By improving analogical reasoning—a key aspect of human intelligence—this work brings us closer to AI that can draw meaningful insights from stories, historical events, or case studies.

Context & Background

  • Analogical reasoning involves drawing parallels between different situations based on structural similarities, a cognitive skill humans use for problem-solving and learning.
  • Large Language Models (LLMs) like GPT-4 have shown impressive capabilities in generating text but often struggle with deep, structured reasoning tasks such as mapping narratives.
  • Previous research in AI has focused on symbolic approaches for analogy, but integrating them with neural models like LLMs remains a challenge due to differences in representation.

What Happens Next

Researchers will likely test this approach on diverse narrative datasets, such as fables, legal cases, or historical accounts, to evaluate its generalization. Further development may integrate these abstractions into educational tools or analytical platforms within 1-2 years. Collaborations between AI labs and domain experts (e.g., in humanities or social sciences) could emerge to refine applications.

Frequently Asked Questions

What is structural mapping in narratives?

Structural mapping involves identifying and aligning the underlying relationships or patterns between different narratives, such as comparing character motivations or plot developments. It goes beyond surface details to find deeper analogies that support reasoning.

How do LLM-derived abstractions help?

LLM-derived abstractions use language models to extract high-level concepts or themes from text, simplifying complex narratives into structured forms. This makes it easier for AI systems to compare and reason across stories, improving accuracy in analogical tasks.

Who benefits from this research?

Educators, AI developers, and analysts benefit, as it could lead to tools for teaching critical thinking or analyzing documents. It also advances AI toward more human-like reasoning, impacting fields like psychology and computational linguistics.

}
Original Source
arXiv:2603.29997v1 Announce Type: cross Abstract: Analogical reasoning is a key driver of human generalization in problem-solving and argumentation. Yet, analogies between narrative structures remain challenging for machines. Cognitive engines for structural mapping are not directly applicable, as they assume pre-extracted entities, whereas LLMs' performance is sensitive to prompt format and the degree of surface similarity between narratives. This gap motivates a key question: What is the impa
Read full article at source

Source

arxiv.org

More from USA

News from Other Countries

🇬🇧 United Kingdom

🇺🇦 Ukraine