Modeling Disclosive Transparency in NLP Application Descriptions
This work provides a quantifiable method for assessing disclosive transparency, which is crucial for researchers and developers to study and improve the clarity of AI system descriptions, especially given potential negative consequences like confusion effects.
The authors introduce neural language model-based probabilistic metrics to model disclosive transparency in NLP application descriptions. They demonstrate that these metrics correlate with user and expert opinions, making them a valid objective proxy for transparency.
Broader disclosive transparency$-$truth and clarity in communication regarding the function of AI systems$-$is widely considered desirable. Unfortunately, it is a nebulous concept, difficult to both define and quantify. This is problematic, as previous work has demonstrated possible trade-offs and negative consequences to disclosive transparency, such as a confusion effect, where "too much information" clouds a reader's understanding of what a system description means. Disclosive transparency's subjective nature has rendered deep study into these problems and their remedies difficult. To improve this state of affairs, We introduce neural language model-based probabilistic metrics to directly model disclosive transparency, and demonstrate that they correlate with user and expert opinions of system transparency, making them a valid objective proxy. Finally, we demonstrate the use of these metrics in a pilot study quantifying the relationships between transparency, confusion, and user perceptions in a corpus of real NLP system descriptions.