ConvAbuse: Data, Analysis, and Benchmarks for Nuanced Abuse Detection in Conversational AI

By Amanda Cercas Curry, Gavin Abercrombie, Verena Rieser. 2021

97 citations recorded by the source corpus.

We present the first English corpus study on abusive language towards three conversational AI systems gathered ‘in the wild’: an open-domain social bot, a rule-based chatbot, and a task-based system. To account for the complexity of the task, we take a more ‘nuanced’ approach where our ConvAI dataset reflects fine-grained notions of abuse, as well as views from multiple expert annotators. We find that the distribution of abuse is vastly different compared to other commonly used datasets, with more sexually tinted aggression towards the virtual persona of these systems. Finally, we report results from bench-marking existing models against this data. Unsurprisingly, we find that there is substantial room for improvement with F1 scores below 90%.

Subfields: Artificial Intelligence, Clinical Psychology, Information Systems, Natural Language Processing, Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Clinical Psychology

Dirk Hovy

In Information Systems

In Natural Language Processing

In Safety, Robustness & Interpretability

Citations in the corpus

Guiding the Release of Safer E2E Conversational AI through Value Sensitive Design (2022, background): “The possibility of adversarial interaction and, more generally, the unpredictability of a system used far outside the training distribution, make it particularly important to not exclusively rely on mitigations such as cleaning up training data to avoid exposing the system to offensive content, as i”
SafetyKit: First Aid for Measuring Safety in Open-domain Conversational Systems (2022, background): “Cercas Curry et al. (2018) try different strategies to deal with abuse directed at their social chat-bot, such as non-sequiturs, appeals to authority, and chastisement.”
Mirages: On Anthropomorphism in Dialogue Systems (2023, background): “This has led to users verbally abusing systems (West et al., 2019), going beyond mere expressions of frustration that one might have with a poorly functioning tool to frequently targeting them with gender-based slurs (Cercas Curry et al., 2021).”