The Inadequacy of Reinforcement Learning From Human Feedback—Radicalizing Large Language Models via Semantic Vulnerabilities
Author:
Affiliation:
1. Cyberoo Pty Ltd., Surrey Hills, NSW, Australia
2. Massey University, Auckland, New Zealand
3. Cyberstronomy Pty Ltd., Ballarat, VIC, Australia
4. RMIT University, Melbourne, VIC, Australia
Publisher
Institute of Electrical and Electronics Engineers (IEEE)
Link
http://xplorestaging.ieee.org/ielx7/7274989/10633860/10474163.pdf?arnumber=10474163
Reference47 articles.
1. Sparks of artificial general intelligence: Early experiments with GPT-4;Bubeck,2023
2. GPT-3: Its Nature, Scope, Limits, and Consequences
3. Recent advances in natural language processing via large pre-trained language models: A survey;Min;ACM Comput. Surveys,2023
4. Large Language Models in Medical Education: Opportunities, Challenges, and Future Directions
5. GPTs are GPTs: An early look at the labor market impact potential of large language models;Eloundou,2023
Cited by 3 articles. 订阅此论文施引文献 订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献
1. The state as a model for AI control and alignment;AI & SOCIETY;2024-09-11
2. Aversion to external feedback suffices to ensure agent alignment;Scientific Reports;2024-09-10
3. From COBIT to ISO 42001: Evaluating cybersecurity frameworks for opportunities, risks, and regulatory compliance in commercializing large language models;Computers & Security;2024-09
1.学者识别学者识别
2.学术分析学术分析
3.人才评估人才评估
"同舟云学术"是以全球学者为主线，采集、加工和组织学术论文而形成的新型学术文献查询和分析系统，可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容，当前同舟云学术共收录了国内外主流学术期刊6万余种，收集的期刊论文及会议论文总量共计约1.5亿篇，并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询！咨询电话：010-8811{复制后删除}0370
www.globalauthorid.com
TOP
Copyright © 2019-2024 北京同舟云网络信息技术有限公司 京公网安备11010802033243号 京ICP备18003416号-3