Tuning Language Models as Training Data Generators for Augmentation-Enhanced Few-Shot Learning

By Yu Meng, Martin Michalski, Jiaxin Huang, Yu Zhang, Tarek F. Abdelzaher, Jiawei Han. 2022

65 citations recorded by the source corpus.

Recent studies have revealed the intriguing few-shot learning ability of pretrained language models (PLMs): They can quickly adapt to a new task when fine-tuned on a small amount of labeled data formulated as prompts, without requiring abundant task-specific annotations. Despite their promising performance, most existing few-shot approaches that only learn from the small training set still underperform fully supervised training by nontrivial margins. In this work, we study few-shot learning with PLMs from a different perspective: We first tune an autoregressive PLM on the few-shot samples and then use it as a generator to synthesize a large amount of novel training samples which augment the original training set. To encourage the generator to produce label-discriminative samples, we train it via weighted maximum likelihood where the weight of each token is automatically adjusted based on a discriminative meta-learning objective. A classification PLM can then be fine-tuned on both the few-shot and the synthetic samples with regularization for better generalization and stability. Our approach FewGen achieves an overall better result across seven classification tasks of the GLUE benchmark than existing few-shot learning methods, improving no-augmentation methods by 5+ average points, and outperforming augmentation methods by 3+ average points.

Subfields: Artificial Intelligence, Large Language Models, Natural Language Processing

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Large Language Models

In Natural Language Processing

Citations in the corpus

Self-Instruct: Aligning Language Models with Self-Generated Instructions (2022, background): “A variety of works have relied on generative LMs for data generation (Schick and Schütze, 2021; Wang et al., 2021; Liu et al., 2022; Meng et al., 2022) or augmentation (Feng et al.”
A Review on Large Language Models: Architectures, Applications, Taxonomies, Open Issues and Challenges (2024, background): “Improving their capacity to generalize and perform well with limited training data is a crucial area of research [174].”
RLAIF vs. RLHF: Scaling Reinforcement Learning from Human Feedback with AI Feedback (2023, methodology): “LLMs have also been extensively used for data generation (Wang et al., 2021b; Meng et al., 2023), augmentation (Feng et al., 2021) and in self-training setups (Wang et al., 2022b; Madaan et al., 2023).”
RLAIF: Scaling Reinforcement Learning from Human Feedback with AI Feedback (2023, unknown): “LLMs have also been extensively used for data generation (Wang et al., 2021b; Meng et al., 2023), augmentation (Feng et al., 2021) and in self-training setups (Wang et al., 2022b; Madaan et al., 2023).”