Unveiling LLMs: The Evolution of Latent Representations in a Temporal Knowledge Graph

arXiv:2404.03623v1 Announce Type: new
Abstract: Large Language Models (LLMs) demonstrate an impressive capacity to recall a vast range of common factual knowledge information. However, unravelling the underlying reasoning of LLMs and explaining their internal mechanisms of exploiting this factual knowledge remain active areas of investigation. Our work analyzes the factual knowledge encoded in the latent representation of LLMs when prompted to assess the truthfulness of factual claims. We propose an end-to-end framework that jointly decodes the factual knowledge embedded in …

abstract arxiv capacity cs.ai cs.cl cs.cy evolution graph however information investigation knowledge knowledge graph language language models large language large language models llms reasoning recall temporal type vast work