Towards Better Few-Shot and Finetuning Performance with Forgetful Causal Language Models

arxiv(2023)

引用 2|浏览132
暂无评分
摘要
Large language models (LLM) trained using the next-token-prediction objective, such as GPT3 and PaLM, have revolutionized natural language processing in recent years by showing impressive zero-shot and few-shot capabilities across a wide range of tasks. In this work, we propose a simple technique that significantly boosts the performance of LLMs without adding computational cost. Our key observation is that, by performing the next token prediction task with randomly selected past tokens masked out, we can improve the quality of the learned representations for downstream language understanding tasks. We hypothesize that randomly masking past tokens prevents over-attending to recent tokens and encourages attention to tokens in the distant past. We find that our method, Forgetful Causal Masking (FCM), significantly improves both few-shot and finetuning performance of PaLM. We further consider a simple extension, T-FCM, which introduces bidirectional context to causal language model without altering the sequence order, and further improves finetuning performance.
更多
查看译文
关键词
forgetful causal
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络