CombLM: Adapting Black-Box Language Models through Small Fine-Tuned Models

Aitor Ormazabal, Mikel Artetxe, Eneko Agirre

Submission Type: Regular Long Paper

Submission Track: Language Modeling and Analysis of Language Models

Keywords: black-box, language models, large language models, adaptation, domain adaptation

TL;DR: We propose a method for adaptation of large black-box models to new domains and tasks through learned combination with fine-tuned small models

Abstract:

Methods for adapting language models (LMs) to new tasks and domains have traditionally assumed white-box access to the model, and work by modifying its parameters.
However, this is incompatible with a recent trend in the field, where the highest quality models are only available as black-boxes through inference APIs. Even when the model weights are available, the computational cost of fine-tuning large LMs can be prohibitive for most practitioners.
In this work, we present a lightweight method for adapting large LMs to new domains and tasks, assuming no access to their weights or intermediate activations. Our approach fine-tunes a small white-box LM and combines it with the large black-box LM at the probability level through a small network, learned on a small validation set. We validate our approach by adapting a large LM (OPT-30B) to several domains and a downstream task (machine translation), observing improved performance in all cases, of up to 9%, while using a domain expert 23x smaller.

Submission Number: 3049