Unpacking Large Language Models with Conceptual Consistency
This work addresses the challenge of evaluating concept understanding in LLMs for researchers and developers, though it is incremental as it builds on existing metrics and datasets.
The authors tackled the problem of assessing whether large language models (LLMs) truly understand concepts by proposing a metric called conceptual consistency, which measures how consistently models respond to queries about related background knowledge; they found that popular LLMs do not necessarily have high conceptual consistency, with significant variation across relations, concepts, and prompts.
If a Large Language Model (LLM) answers "yes" to the question "Are mountains tall?" then does it know what a mountain is? Can you rely on it responding correctly or incorrectly to other questions about mountains? The success of Large Language Models (LLMs) indicates they are increasingly able to answer queries like these accurately, but that ability does not necessarily imply a general understanding of concepts relevant to the anchor query. We propose conceptual consistency to measure a LLM's understanding of relevant concepts. This novel metric measures how well a model can be characterized by finding out how consistent its responses to queries about conceptually relevant background knowledge are. To compute it we extract background knowledge by traversing paths between concepts in a knowledge base and then try to predict the model's response to the anchor query from the background knowledge. We investigate the performance of current LLMs in a commonsense reasoning setting using the CSQA dataset and the ConceptNet knowledge base. While conceptual consistency, like other metrics, does increase with the scale of the LLM used, we find that popular models do not necessarily have high conceptual consistency. Our analysis also shows significant variation in conceptual consistency across different kinds of relations, concepts, and prompts. This serves as a step toward building models that humans can apply a theory of mind to, and thus interact with intuitively.