paper-with-me

홈 › Papers

Benchmark Early and Red Team Often: A Framework for Assessing and Managing Dual-Use Hazards of AI Foundation Models

2024-05-15 · Anthony M. Barrett, Krystal Jackson, Evan R. Murphy, Nada Madkour, Jessica Newman

A concern about cutting-edge or "frontier" AI foundation models is that an adversary may use the models for preparing chemical, biological, radiological, nuclear, (CBRN), cyber, or other attacks. At least two methods can identify foundation models with potential dual-use capability; each has advantages and disadvantages: A. Open benchmarks (based on openly available questions and answers), which are low-cost but accuracy-limited by the need to omit security-sensitive details; and B. Closed red team evaluations (based on private evaluation by CBRN and cyber experts), which are higher-cost but can achieve higher accuracy by incorporating sensitive details. We propose a research and risk-management approach using a combination of methods including both open benchmarks and closed red team evaluations, in a way that leverages advantages of both methods. We recommend that one or more groups of researchers with sufficient resources and access to a range of near-frontier and frontier foundation models run a set of foundation models through dual-use capability evaluation benchmarks and red team evaluations, then analyze the resulting sets of models' scores on benchmark and red team evaluations to see how correlated those are. If, as we expect, there is substantial correlation between the dual-use potential benchmark scores and the red team evaluation scores, then implications include the following: The open benchmarks should be used frequently during foundation model development as a quick, low-cost measure of a model's dual-use potential; and if a particular model gets a high score on the dual-use potential benchmark, then more in-depth red team assessments of that model's dual-use capability should be performed. We also discuss limitations and mitigations for our approach, e.g., if model developers try to game benchmarks by including a version of benchmark test data in a model's training data.

📄 PDF Abstract BibTeX arXiv:2405.10986

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

AutoRedTeamer: Autonomous Red Teaming with Lifelong Attack Integration

2025-03-20 · Andy Zhou, Kevin Wu, Francesco Pinto, Zhaorun Chen 외

As large language models (LLMs) become increasingly capable, security and safety evaluation are crucial. While current red teaming approaches have made strides in assessing LLM vulnerabilities, they often rely heavily on…

Red Teaming

Embodied Red Teaming for Auditing Robotic Foundation Models

2024-11-27 · Sathwik Karnik, Zhang-Wei Hong, Nishant Abhangi, Yen-Chen Lin 외

Language-conditioned robot models have the potential to enable robots to perform a wide range of tasks based on natural language instructions. However, assessing their safety and effectiveness remains challenging because…

Red Teaming

Assessing Risks of Large Language Models in Mental Health Support: A Framework for Automated Clinical AI Red Teaming

2026-02-23 · Ian Steenstra, Paola Pedrelli, Weiyan Shi, Stacy Marsella 외 arxiv

Large Language Models (LLMs) are increasingly utilized for mental health support; however, current safety benchmarks often fail to detect the complex, longitudinal risks inherent in therapeutic dialogue. We introduce an …

Red Teaming

Quantifying the Cognitive Extent of Science

2015-10-30 · Staša Milojević

While the modern science is characterized by an exponential growth in scientific literature, the increase in publication volume clearly does not reflect the expansion of the cognitive boundaries of science. Nevertheless,…

ArticlesAstronomy

Red Teaming LLMs as Socio-Technical Practice: From Exploration and Data Creation to Evaluation

2026-02-10 · Adriana Alvarado Garcia, Ruyuan Wan, Ozioma C. Oguine, Karla Badillo-Urquiola arxiv

Recently, red teaming, with roots in security, has become a key evaluative approach to ensure the safety and reliability of Generative Artificial Intelligence. However, most existing work emphasizes technical benchmarks …

Red Teaming