Matches in SemOpenAlex for { <https://semopenalex.org/work/W4385571796> ?p ?o ?g. }
Showing items 1 to 55 of
55
with 100 items per page.
- W4385571796 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%." @default.
- W4385571796 created "2023-08-05" @default.
- W4385571796 creator A5039633744 @default.
- W4385571796 creator A5050195037 @default.
- W4385571796 creator A5068296767 @default.
- W4385571796 date "2023-01-01" @default.
- W4385571796 modified "2023-09-27" @default.
- W4385571796 title "Foveate, Attribute, and Rationalize: Towards Physically Safe and Trustworthy AI" @default.
- W4385571796 doi "https://doi.org/10.18653/v1/2023.findings-acl.701" @default.
- W4385571796 hasPublicationYear "2023" @default.
- W4385571796 type Work @default.
- W4385571796 citedByCount "0" @default.
- W4385571796 crossrefType "proceedings-article" @default.
- W4385571796 hasAuthorship W4385571796A5039633744 @default.
- W4385571796 hasAuthorship W4385571796A5050195037 @default.
- W4385571796 hasAuthorship W4385571796A5068296767 @default.
- W4385571796 hasBestOaLocation W43855717961 @default.
- W4385571796 hasConcept C112930515 @default.
- W4385571796 hasConcept C143299363 @default.
- W4385571796 hasConcept C144133560 @default.
- W4385571796 hasConcept C151730666 @default.
- W4385571796 hasConcept C153701036 @default.
- W4385571796 hasConcept C15744967 @default.
- W4385571796 hasConcept C2779343474 @default.
- W4385571796 hasConcept C38652104 @default.
- W4385571796 hasConcept C41008148 @default.
- W4385571796 hasConcept C77805123 @default.
- W4385571796 hasConcept C86803240 @default.
- W4385571796 hasConceptScore W4385571796C112930515 @default.
- W4385571796 hasConceptScore W4385571796C143299363 @default.
- W4385571796 hasConceptScore W4385571796C144133560 @default.
- W4385571796 hasConceptScore W4385571796C151730666 @default.
- W4385571796 hasConceptScore W4385571796C153701036 @default.
- W4385571796 hasConceptScore W4385571796C15744967 @default.
- W4385571796 hasConceptScore W4385571796C2779343474 @default.
- W4385571796 hasConceptScore W4385571796C38652104 @default.
- W4385571796 hasConceptScore W4385571796C41008148 @default.
- W4385571796 hasConceptScore W4385571796C77805123 @default.
- W4385571796 hasConceptScore W4385571796C86803240 @default.
- W4385571796 hasLocation W43855717961 @default.
- W4385571796 hasOpenAccess W4385571796 @default.
- W4385571796 hasPrimaryLocation W43855717961 @default.
- W4385571796 hasRelatedWork W1967906317 @default.
- W4385571796 hasRelatedWork W2028906177 @default.
- W4385571796 hasRelatedWork W2040955267 @default.
- W4385571796 hasRelatedWork W2075994112 @default.
- W4385571796 hasRelatedWork W2160853337 @default.
- W4385571796 hasRelatedWork W2363664336 @default.
- W4385571796 hasRelatedWork W2767109334 @default.
- W4385571796 hasRelatedWork W2902546803 @default.
- W4385571796 hasRelatedWork W3209560528 @default.
- W4385571796 hasRelatedWork W4313569620 @default.
- W4385571796 isParatext "false" @default.
- W4385571796 isRetracted "false" @default.
- W4385571796 workType "article" @default.