The Effect of Scaling, Retrieval Augmentation and Form on the Factual Consistency of Language Models

• Large Language Models (LLMs) make natural interfaces to factual knowledge, but their usefulness is limited by their tendency to deliver inconsistent answers to semantically equivalent questions. For example, a model might supply the answer “Edinburgh” to “Anne Redpath passed away in X.” and “London” to “Anne Redpath’s life ended in X.” In this work, we identify potential causes of inconsistency and evaluate the effectiveness of two mitigation strategies: up-scaling and augmenting the LM with a passage retrieval database. Our results on the LLaMA and Atlas models show that both strategies reduce inconsistency but that retrieval augmentation is considerably more efficient. We further consider and disentangle the consistency contributions of different components of Atlas. For all LMs evaluated we find that syntactical form and task artifacts impact consistency. Taken together, our results provide a better understanding of the factors affecting the factual consistency of language models.

Subject headings

NATURVETENSKAP  -- Data- och informationsvetenskap -- Språkteknologi (hsv//swe)

NATURAL SCIENCES  -- Computer and Information Sciences -- Language Technology (hsv//eng)

HUMANIORA  -- Språk och litteratur -- Studier av enskilda språk (hsv//swe)

HUMANITIES  -- Languages and Literature -- Specific Languages (hsv//eng)

Keyword

language models

LLMs

retrieval

consistency

accuracy

evaluation

knowledge representation

Publication and Content Type

ref (subject category)

kon (subject category)