The advent of advanced AI underscores the urgent need for comprehensive
safety evaluations, necessitating collaboration across communities (i.e., AI,
software engineering, and governance). However, divergent practices and
terminologies across these communities, combined with the complexity of AI
systems-of which models are only a part-and environmental affordances (e.g.,
access to tools), obstruct effective communication and comprehensive
evaluation. This paper proposes a framework for AI system evaluation comprising
three components: 1) harmonised terminology to facilitate communication across
communities involved in AI safety evaluation; 2) a taxonomy identifying
essential elements for AI system evaluation; 3) a mapping between AI lifecycle,
stakeholders, and requisite evaluations for accountable AI supply chain. This
framework catalyses a deeper discourse on AI system evaluation beyond
model-centric approaches.

该研究提出了一个 AI 系统评估的框架，包括协调术语、鉴定要素和建立映射，以促进跨学科的合作和综合评估。

推进人工智能安全的人工智能系统评估框架：术语、分类、生命周期映射

An AI System Evaluation Framework for Advancing AI Safety: Terminology,  Taxonomy, Lifecycle Mapping

Pretrained language models underpin several AI applications, but their high
computational cost for training limits accessibility. Initiatives such as BLOOM
and StarCoder aim to democratize access to pretrained models for collaborative
community development. However, such existing models face challenges: limited
multilingual capabilities, continual pretraining causing catastrophic
forgetting, whereas pretraining from scratch is computationally expensive, and
compliance with AI safety and development laws. This paper presents Aurora-M, a
15B parameter multilingual open-source model trained on English, Finnish,
Hindi, Japanese, Vietnamese, and code. Continually pretrained from
StarCoderPlus on 435 billion additional tokens, Aurora-M surpasses 2 trillion
tokens in total training token count. It is the first open-source multilingual
model fine-tuned on human-reviewed safety instructions, thus aligning its
development not only with conventional red-teaming considerations, but also
with the specific concerns articulated in the Biden-Harris Executive Order on
the Safe, Secure, and Trustworthy Development and Use of Artificial
Intelligence. Aurora-M is rigorously evaluated across various tasks and
languages, demonstrating robustness against catastrophic forgetting and
outperforming alternatives in multilingual settings, particularly in safety
evaluations. To promote responsible open-source LLM development, Aurora-M and
its variants are released at
this https URL .

Aurora-M is a multilingual open-source pretrained language model that addresses challenges such as limited multilingual capabilities and catastrophic forgetting, surpassing 2 trillion training tokens. It aligns with safety considerations and outperforms alternatives in multilingual settings.