ConvXAI: a System for Multimodal Interaction with Any Black-box Explainer

Malandri, Lorenzo; Mercorio, Fabio; Mezzanzanica, Mario; Nobani, Navid

doi:10.1007/s12559-022-10067-7

Cited by 13 publications

(16 citation statements)

References 68 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This capability aids in addressing vague follow-up queries. Moreover, Malandri et al [52] combine rule-based Dialogue State Tracking with a tailored Dialogue Policy module, facilitating dynamic response decisions based on current context and conversation state.…”

Section: Dialogue Capabilitiesmentioning

confidence: 99%

“…To address RQ1.8, we conducted a detailed analysis of studies to identify indications of whether and to what extent the systems have the recommended properties. Our analysis reveals the emergence of approaches utilizing drill-down dialogues [15] and clarification dialogues [52] that allow the user to clarify or argue about an explanation offered by the system. Beyond these properties, however, the other recommended properties have yet to be implemented.…”

Section: Desirable Properties Suggested From Theoretical Workmentioning

confidence: 99%

“…Furthermore, both Sklar and Azhar [42] and Hernandez-Bocanegra and Ziegler [51] advocate for argumentationbased dialogues, fostering dynamic exchanges where users can question the system's suggestions. Inspired by the protocol of Madumal et al [70], Malandri et al [52] incorporate the possibility of clarification dialogues into their dialogue system, allowing users to question the explanation provided.…”

Section: Arguing About Explanationsmentioning

confidence: 99%

“…Each group benefits from tailored conversational explanations to enhance their understanding and interaction with the AI systems. Similarly, Malandri et al [52] designed a system that adapts responses to the user knowledge, which is accessed through initial dialogue questions that help differentiate between the profiles end-user, developer, and manager.…”

Section: Targeting Multiple User Groupsmentioning

confidence: 99%

“…Rather than using existing solutions, Sklar and Azhar [42], Viros Martin and Selva [45], Malandri et al [52] and Hernandez-Bocanegra and Ziegler [51] take a different approach by crafting custom user interfaces with chat windows, dependencies and registrations to external services.…”

Section: Communication and User Interfacementioning

confidence: 99%

See 4 more Smart Citations

Beyond One-Shot Explanations: A Systematic Literature Review of Dialogue-Based XAI Approaches

Mindlin,

Beer,

Sieger

et al. 2024

Preprint

View full text Add to dashboard Cite

In the last decade, there has been increasing interest in allowing users to understand how the predictions of machine-learned models come about, thus increasing transparency and empowering users to understand and potentially contest those decisions.Dialogue-based approaches, in contrast to traditional one-shot eXplainable Artificial Intelligence (XAI) methods, facilitate interactive, in-depth exploration through multi-turn dialogues, simulating expert conversations. This paper reviews the current state of dialogue-based XAI, presenting a systematic review of 1,339 publications, narrowed down to 14 based on inclusion criteria. We explore theoretical foundations of the systems, propose key dimensions along which different solutions to dialogue-based XAI differ, and identify key use cases, target audiences, system components, and the types of supported queries and responses. Furthermore, we investigate the current paradigms by which systems are evaluated and highlight their key limitations. Key findings include identifying the main use cases, objectives, and audiences targeted by dialogue-based XAI methods, and summarize the main types of questions and information needs. Beyond discussing avenues for future work, we present a meta-architecture for these systems from existing literature and outlined prevalent theoretical frameworks.

show abstract

Section: Dialogue Capabilitiesmentioning

confidence: 99%

Section: Desirable Properties Suggested From Theoretical Workmentioning

confidence: 99%