analiza aiqbench
This study challenges the notion that human reasoning relies on abstract world models by demonstrating that both humans and LLMs exhibit similar pattern-matching behaviors in everyday causal tasks. By isolating specific attention heads, researchers show these mechanisms can predict errors driven by irrelevant prompt details across biological and artificial systems alike. The findings suggest a fundamental convergence between how people and machines process common-sense logic, reframing "reasoning" as largely statistical association rather than principled abstraction.
Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning
arXiv:2606.13607v2 Announce Type: replace Abstract: When large language models (LLMs) fail to generalize or make haphazard errors in reasoning, it is often taken as evidence that LLMs are not truly reasoning, but rather performing a kind of pattern matching. The implication is that people's behavior does not exhibit the same types of failures because human reasoning uses principled and abstract world models. We evaluate human participants and 25 LLMs on their ability to engage in common-sense reasoning about a variety of everyday situations and observe similar patterns of errors in both people and models. We then identify the set of attention heads driving LLM responses and find that these heads implement a form of pattern-matching. These attention heads allow us to predict seemingly inexplicable reasoning errors in people caused by ostensibly irrelevant prompt details. Taken together, our results suggest that everyday causal reasoning in people and LLMs is more consistent with a form of pattern-matching than with abstract world models.
Abstract:When large language models (LLMs) fail to generalize or make haphazard errors in reasoning, it is often taken as evidence that LLMs are not truly reasoning, but rather performing a kind of pattern matching. The implication is that people's behavior does not exhibit the same types of failures because human reasoning uses principled and abstract world models. We evaluate human participants and 25 LLMs on their ability to engage in common-sense reasoning about a variety of everyday situations and observe similar patterns of errors in both people and models. We then identify the set of attention heads driving LLM responses and find that these heads implement a form of pattern-matching. These attention heads allow us to predict seemingly inexplicable reasoning errors in people caused by ostensibly irrelevant prompt details. Taken together, our results suggest that everyday causal reasoning in people and LLMs is more consistent with a form of pattern-matching than with abstract world models.
Artificial Intelligence (cs.AI)
arXiv:2606.13607 [cs.AI]
(or arXiv:2606.13607v2 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2606.13607 Focus to learn more arXiv-issued DOI via DataCite
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 247 cuvinte
Subiecte conexe
Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning
arXiv · 2 oct. 2026
Reasoning as Pattern Matching: Shared Mechanisms in Human and LLM Everyday Reasoning
arXiv · 12 iun. 2026
Game Design Creativity evals
LessWrong · 12 oct. 2026
The Cost of Utopias (a Dialog)
LessWrong · 18 sept. 2026