Language Models are Few-Shot Learners

计算机科学 任务(项目管理) 语言模型 自然语言处理 判决 人工智能 词(群论) 简单(哲学) 语言学 认识论 哲学 经济 管理
作者
Brown, Tom B.,Benjamin Mann,Nick Ryder,Melanie Subbiah,Jared Kaplan A,Prafulla Dhariwal,Arvind Neelakantan,Pranav Shyam,Girish Sastry,Amanda Askell,Sandhini Agarwal,Ariel Herbert-Voss,Gretchen Marie Krueger,Tom Henighan,Rewon Child,Ramesh, Aditya,Ziegler, Daniel M.,Jeffrey Wu,Clemens Winter,Christopher Hesse
出处
期刊:Cornell University - arXiv [Cornell University]
卷期号:33: 1877-1901 被引量:16451
标识
DOI:10.48550/arxiv.2005.14165
摘要

Recent work has demonstrated substantial gains on many NLP tasks and benchmarks by pre-training on a large corpus of text followed by fine-tuning on a specific task. While typically task-agnostic in architecture, this method still requires task-specific fine-tuning datasets of thousands or tens of thousands of examples. By contrast, humans can generally perform a new language task from only a few examples or from simple instructions - something which current NLP systems still largely struggle to do. Here we show that scaling up language models greatly improves task-agnostic, few-shot performance, sometimes even reaching competitiveness with prior state-of-the-art fine-tuning approaches. Specifically, we train GPT-3, an autoregressive language model with 175 billion parameters, 10x more than any previous non-sparse language model, and test its performance in the few-shot setting. For all tasks, GPT-3 is applied without any gradient updates or fine-tuning, with tasks and few-shot demonstrations specified purely via text interaction with the model. GPT-3 achieves strong performance on many NLP datasets, including translation, question-answering, and cloze tasks, as well as several tasks that require on-the-fly reasoning or domain adaptation, such as unscrambling words, using a novel word in a sentence, or performing 3-digit arithmetic. At the same time, we also identify some datasets where GPT-3's few-shot learning still struggles, as well as some datasets where GPT-3 faces methodological issues related to training on large web corpora. Finally, we find that GPT-3 can generate samples of news articles which human evaluators have difficulty distinguishing from articles written by humans. We discuss broader societal impacts of this finding and of GPT-3 in general.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
陈辉发布了新的文献求助10
1秒前
1秒前
果粒橙自由版完成签到 ,获得积分10
2秒前
梅思寒完成签到 ,获得积分10
3秒前
AZQ完成签到,获得积分10
3秒前
4秒前
qin202569发布了新的文献求助10
4秒前
欢呼致远发布了新的文献求助10
5秒前
极限001的应助被卡卡西西西采纳,获得20
6秒前
充电宝的应助被巴拉阿拉采纳,获得30
7秒前
lw完成签到 ,获得积分10
8秒前
炙热冰完成签到,获得积分10
8秒前
9秒前
鸡鱼蚝发布了新的文献求助10
9秒前
9秒前
10秒前
杨乐完成签到,获得积分10
12秒前
kdxjlm发布了新的文献求助10
15秒前
湫殇发布了新的文献求助20
15秒前
郁文完成签到,获得积分10
15秒前
perry完成签到,获得积分10
16秒前
Ava的应助被小唐采纳,获得10
17秒前
17秒前
名井南来北往完成签到,获得积分10
17秒前
17秒前
18秒前
18秒前
早睡早起完成签到,获得积分10
19秒前
20秒前
21秒前
巴拉阿拉发布了新的文献求助30
23秒前
23秒前
qin202569发布了新的文献求助10
24秒前
24秒前
kdxjlm完成签到,获得积分10
25秒前
decimalpoint完成签到,获得积分10
25秒前
26秒前
叶95发布了新的文献求助20
26秒前
田様的应助被沉默笑蓝采纳,获得10
27秒前
大模型的应助被XPR采纳,获得10
28秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
奈特基础组织学:配套组织病理学(第4版)(Netter's Essential Histology With Correlated Histopathology) 800
八旗制度の研究 600
Boundaries and Bodhisattvas 500
高分子基功能界面水蒸发材料与应用研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8181767
求助须知:如何正确求助?哪些是违规求助? 9571480
关于积分的说明 21818225
捐赠科研通 7778030
什么是DOI,文献DOI怎么找? 3433524
关于科研通互助平台的介绍 2557696
邀请新用户注册赠送积分活动 2479598