Recursive Knowledge Acquisition in Language Models

Home

Discover

Bookmarks

Keep Up With Latest Trending Papers. Computer Science, AI and more.Subscribe

Added to collectionsTop 100 Papers By Signal Trends and Top 100 Trending AI & Computer Science Papers

CC BY

Source arXiv

Machine Learning

Artificial Intelligence

Computation and Language

Recursive Knowledge Acquisition in Language Models

Memorizing Transformers

Language models typically need to be trained or finetuned in order to acquire new knowledge, which involves updating their weights.

We instead envision language models that can simply read and memorize new data at inference time, thus acquiring new knowledge immediately.

In this work, we extend languagemodels with the ability to memorize the internal representations of past inputs.

We demonstrate that an approximate knn lookup into a non-differentiable memory of recent (key, value) pairs improves language modeling across various benchmarks and tasks, including generic webtext (c4), math papers (arxiv), books (pg-19), code (github), as well as formal theorems (isabelle).

We show that the performance steadily improves when we increase the size of memory upto 262k tokens.

On benchmarks including code and mathematics, we find that the model is capable of making use of newly defined functions and theorems during test time.

Authors

Yuhuai Wu, Markus N. Rabe, DeLesley Hutchins, Christian Szegedy

◐ Recommended Follows

◐ Latest Activity