We present QLoRA, an efficient finetuning approach that reduces memory usage enough to
finetune a 65B parameter model on a single 48GB GPU while preserving full 16-bit …

Few-shot in-context learning (ICL) enables pre-trained language models to perform a
previously-unseen task without any gradient-based training by feeding a small number of …

Prompting has shown impressive success in enabling large pretrained language models
(LMs) to perform diverse NLP tasks, especially when only few downstream data are …

Code-generating large language models map natural language to code. However, only a
small portion of the infinite space of naturalistic utterances is effective at guiding code …

Abstract We introduce Directional Stimulus Prompting, a novel framework for guiding black-
box large language models (LLMs) towards specific desired outputs. Instead of directly …

Low-rank adaptations (LoRA) are often employed to fine-tune large language models
(LLMs) for new tasks. This paper investigates LoRA composability for cross-task …

In this paper, we explore the challenging problem of performing a generative task in a target
language when labeled data is only available in English, using summarization as a case …

With the continuous growth of large language models, the process of fine-tuning these
models for new tasks has become increasingly parameter-intensive. Prompt tuning, a …

Aligning language models (LMs) with preferences is an important problem in natural
language generation. A key challenge is that preferences are typically provided at the …

Fine-tuning large language models is becoming ever more impractical due to their rapidly-
growing scale. This motivates the use of parameter-efficient adaptation methods such as …