Pre-training Limited Memory Language Models with Internal and External Knowledge

Linxi Zhao, Sofian Zalouk, Christian Belardi, Justin Lovelace, Jin Peng Zhou, Ryan Thomas Noonan, Dongyoung Go, Kilian Q Weinberger, Yoav Artzi, Jennifer J. Sun

Keywords: Pretrained Large Language Models, Knowledge Offloading

TL;DR: A new class of language models that offloads factual knowledge to an external database rather than encoding it in their parameters

Abstract: Neural language models are black-boxes--both linguistic patterns and factual knowledge are distributed across billions of opaque parameters. This entangled encoding makes it difficult to reliably inspect, verify, or update specific facts. We introduce Limited Memory Language Models (LMLM), a new class of language models that externalizes factual knowledge to external database during pre-training rather than memorizing them. Our pre-training approach strategically masks externally retrieved factual values from the training loss, thereby teaching the model to perform targeted lookups rather than relying on memorization in model weights. Our experiments demonstrate that LMLMs achieve competitive performance compared to significantly larger LLMs on standard benchmarks, while offering the advantages of explicit, editable, and verifiable knowledge bases.

Primary Area: foundation or frontier models, including LLMs

Submission Number: 1787