Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI

Alex Mei, Sharon Levy, William Yang Wang


Abstract
Users’ physical safety is an increasing concern as the market for intelligent systems continues to grow, where unconstrained systems may recommend users dangerous actions that can lead to serious injury. Covertly unsafe text is an area of particular interest, as such text may arise from everyday scenarios and are challenging to detect as harmful. We propose FARM, a novel framework leveraging external knowledge for trustworthy rationale generation in the context of safety. In particular, FARM foveates on missing knowledge to qualify the information required to reason in specific scenarios and retrieves this information with attribution to trustworthy sources. This knowledge is used to both classify the safety of the original text and generate human-interpretable rationales, shedding light on the risk of systems to specific user groups and helping both stakeholders manage the risks of their systems and policymakers to provide concrete safeguards for consumer safety. Our experiments show that FARM obtains state-of-the-art results on the SafeText dataset, showing absolute improvement in safety classification accuracy by 5.9%.
Anthology ID:
2023.findings-acl.701
Volume:
Findings of the Association for Computational Linguistics: ACL 2023
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Anna Rogers, Jordan Boyd-Graber, Naoaki Okazaki
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
11021–11036
Language:
URL:
https://aclanthology.org/2023.findings-acl.701
DOI:
10.18653/v1/2023.findings-acl.701
Bibkey:
Cite (ACL):
Alex Mei, Sharon Levy, and William Yang Wang. 2023. Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI. In Findings of the Association for Computational Linguistics: ACL 2023, pages 11021–11036, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI (Mei et al., Findings 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.findings-acl.701.pdf
Video:
 https://aclanthology.org/2023.findings-acl.701.mp4