跳到正文
返回论文列表
cs.AI提交于 已译

基础模型的机会与风险

On the Opportunities and Risks of Foundation Models

Rishi Bommasani · Drew A. Hudson · Ehsan Adeli · Russ Biagio Altman · Simran Arora · Sydney von Arx · et al.

中文摘要

随着BERT、DALL-E、GPT-3等模型的出现,AI正经历范式转变:这些模型在大规模广泛数据上训练,可适应广泛下游任务。我们称这些模型为基础模型(foundation models),以强调其关键中心但尚不完整的特性。本报告全面阐述了基础模型的机会与风险,涵盖其能力(如语言、视觉、机器人、推理、人机交互)和技术原理(如模型架构、训练过程、数据、系统、安全、评估、理论),以及应用(如法律、医疗、教育)和社会影响(如不平等、滥用、经济和环境影响、法律与伦理考量)。尽管基础模型基于标准深度学习和迁移学习,但其规模带来了新的涌现能力,且跨众多任务的有效性推动了同质化(homogenization)。同质化提供强大杠杆但需谨慎,因为基础模型的缺陷会被所有下游适配模型继承。尽管基础模型即将广泛部署,但由于其涌现特性,我们目前对其工作原理、失败时机甚至能力范围缺乏清晰理解。为解决这些问题,我们认为基础模型的许多关键研究需要与其根本的社会技术性质相匹配的深度跨学科合作。

关键要点

  1. 01问题:基础模型虽广泛部署,但对其工作原理、失败条件和能力范围缺乏清晰理解。
  2. 02方法:报告系统梳理基础模型的能力、技术原理、应用及社会影响,强调跨学科合作。
  3. 03结果:基础模型基于标准深度学习和迁移学习,规模带来涌现能力,但同质化导致缺陷被下游继承。
  4. 04局限:基础模型的涌现特性使当前理解不足,需社会技术视角的深入研究。

解读

尚无解读。

原始英文摘要

AI is undergoing a paradigm shift with the rise of models (e.g., BERT, DALL-E, GPT-3) that are trained on broad data at scale and are adaptable to a wide range of downstream tasks. We call these models foundation models to underscore their critically central yet incomplete character. This report provides a thorough account of the opportunities and risks of foundation models, ranging from their capabilities (e.g., language, vision, robotics, reasoning, human interaction) and technical principles(e.g., model architectures, training procedures, data, systems, security, evaluation, theory) to their applications (e.g., law, healthcare, education) and societal impact (e.g., inequity, misuse, economic and environmental impact, legal and ethical considerations). Though foundation models are based on standard deep learning and transfer learning, their scale results in new emergent capabilities,and their effectiveness across so many tasks incentivizes homogenization. Homogenization provides powerful leverage but demands caution, as the defects of the foundation model are inherited by all the adapted models downstream. Despite the impending widespread deployment of foundation models, we currently lack a clear understanding of how they work, when they fail, and what they are even capable of due to their emergent properties. To tackle these questions, we believe much of the critical research on foundation models will require deep interdisciplinary collaboration commensurate with their fundamentally sociotechnical nature.

同方向论文 · cs.AI

查看全部 →

反向发现 · 这篇论文相关的工具与提示词

去交叉发现 →

关联工具 · 2

The Decoder85

Anthropic 发布 Claude Fable 5.2 与 Mythos 5.2,IPO 前重回基准榜首

Anthropic 发布 Claude Fable 5.2 与 Mythos 5.2,两者共享同一模型底座,区别在于访问权限:Fable 面向所有用户并屏蔽网络安全和生物领域的风险任务,Mythos 解除部分限制,仅向通过 Anthropic 网络安全或生命科学验证计划的组织开放。

推荐理由:材料给出 Fable 5.2 与 Mythos 5.2 的基准排名、定价和受限开放方式,可对照其与 Opus 5.5 的取舍。

OpenAI News78

OpenAI 发布 GPT-6 Cyber 网络安全模型

OpenAI 发布网络安全模型 GPT-6 Cyber,用于发现可利用漏洞并开发修复方案。该模型经过训练,被要求只在授权范围内行动。

推荐理由:OpenAI 推出面向网络安全的专用模型,强调在授权范围内工作,可观察安全智能体的对齐约束如何落地。