FreeLM: Fine-Tuning-Free Language Model
This addresses the problem of resource-intensive fine-tuning for NLP practitioners, offering a more efficient alternative, though it appears incremental as it builds on existing pre-training paradigms.
The paper tackles the high deployment costs and low training efficiency of fine-tuning pre-trained language models by introducing a fine-tuning-free strategy that incorporates both language and teacher signals, resulting in FreeLM, a 0.3B parameter model that outperforms larger models like GPT-3 and InstructGPT on various language understanding tasks.
Pre-trained language models (PLMs) have achieved remarkable success in NLP tasks. Despite the great success, mainstream solutions largely follow the pre-training then finetuning paradigm, which brings in both high deployment costs and low training efficiency. Nevertheless, fine-tuning on a specific task is essential because PLMs are only pre-trained with language signal from large raw data. In this paper, we propose a novel fine-tuning-free strategy for language models, to consider both language signal and teacher signal. Teacher signal is an abstraction of a battery of downstream tasks, provided in a unified proposition format. Trained with both language and strong task-aware teacher signals in an interactive manner, our FreeLM model demonstrates strong generalization and robustness. FreeLM outperforms large models e.g., GPT-3 and InstructGPT, on a range of language understanding tasks in experiments. FreeLM is much smaller with 0.3B parameters, compared to 175B in these models.