Memorization $\neq$ Understanding: Do Large Language Models Have the Ability of Scenario Cognition?
This work addresses a critical limitation in LLMs' semantic understanding for NLP applications, though it is incremental in exposing existing model weaknesses rather than proposing new solutions.
The researchers investigated whether large language models (LLMs) achieve generalization through deep semantic understanding or mere memorization by evaluating their scenario cognition ability, finding that current LLMs predominantly rely on superficial memorization and fail at robust semantic scenario cognition even in simple cases.
Driven by vast and diverse textual data, large language models (LLMs) have demonstrated impressive performance across numerous natural language processing (NLP) tasks. Yet, a critical question persists: does their generalization arise from mere memorization of training data or from deep semantic understanding? To investigate this, we propose a bi-perspective evaluation framework to assess LLMs' scenario cognition - the ability to link semantic scenario elements with their arguments in context. Specifically, we introduce a novel scenario-based dataset comprising diverse textual descriptions of fictional facts, annotated with scenario elements. LLMs are evaluated through their capacity to answer scenario-related questions (model output perspective) and via probing their internal representations for encoded scenario elements-argument associations (internal representation perspective). Our experiments reveal that current LLMs predominantly rely on superficial memorization, failing to achieve robust semantic scenario cognition, even in simple cases. These findings expose critical limitations in LLMs' semantic understanding and offer cognitive insights for advancing their capabilities.