kth.sePublications KTH
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
(Mis)Communicating with our AI Systems
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Speech, Music and Hearing, TMH. (MUSAiC)ORCID iD: 0000-0003-1098-6873
KTH, School of Electrical Engineering and Computer Science (EECS), Intelligent systems, Speech, Music and Hearing, TMH.ORCID iD: 0000-0003-2549-6367
2025 (English)In: Proceedings of the 2025 CHI Conference on Human Factors in Computing Systems, New York, NY, USA: Association for Computing Machinery (ACM) , 2025, article id 416Conference paper, Published paper (Refereed)
Abstract [en]

Explainable Artificial Intelligence (XAI) is a discipline concerned with understanding predictions of AI systems. What is ultimately desired from XAI methods is for an AI system to link its input and output in a way that is interpretable with reference to the environment in which it is applied. A variety of methods have been proposed, but we argue in this paper that what has yet to be considered is miscommunication: the failure to convey and/or interpret an explanation accurately. XAI can be seen as a communication process and thus looking at how humans explain things to each other can provide guidance to its application and evaluation. We motivate a specific model of communication to help identify essential components of the process, and show the critical importance for establishing common ground, i.e., shared mutual knowledge, beliefs, and assumptions of the participants communicating.

Place, publisher, year, edition, pages
New York, NY, USA: Association for Computing Machinery (ACM) , 2025. article id 416
Keywords [en]
Communication, Miscommunication, Dialog, Mutual-Understanding, Conversation, Explanation, Explainability, Explainable AI
National Category
Computer Sciences
Research subject
Computer Science
Identifiers
URN: urn:nbn:se:kth:diva-363375DOI: 10.1145/3706598.3713771OAI: oai:DiVA.org:kth-363375DiVA, id: diva2:1958494
Conference
CHI 2025: CHI Conference on Human Factors in Computing Systems, Yokohama, Japan, 26 April - 1 May, 2025
Note

Part of ISBN 9798400713941

QC 20250515

Available from: 2025-05-15 Created: 2025-05-15 Last updated: 2025-11-05Bibliographically approved
In thesis
1. Perspectives on AI and Music: Representation, Detection, and Explanation in the Age of AI-Generated Music
Open this publication in new window or tab >>Perspectives on AI and Music: Representation, Detection, and Explanation in the Age of AI-Generated Music
2025 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

The development and commercial exploitation of AI systems for generating music is impacting how music is created, distributed, and used. This thesis addresses three pressing technical challenges posed by these impacts: (1) how to develop systematic methods for evaluating and representing AI-generated music, including statistical tests and melodic analysis; (2) how to define and build music AI detection systems, revealing vulnerabilities to platform bias and audio manipulation; and (3) how to reframe Explainable AI (XAI) as a communication problem, identifying misalignment risks between AI explanations and human interpretation. The results contribute practical methods for music analysis and detection using AI, and highlight key limitations in current approaches. Overall, this thesis lays a foundation for future work on robust evaluation frameworks and improved explanation methods tailored for music AI systems.

Abstract [sv]

Utvecklingen och det kommersiella utnyttjandet av AI-system för att generera musik påverkar hur musik skapas, distribueras och används. Denna avhandling tar upp tre angelägna tekniska utmaningar som dessa effekter medför: (1) hur man utvecklar systematiska metoder för att utvärdera och representera AI-genererad musik, inklusive statistiska tester och melodisk analys; (2) hur man definierar och bygger AI-detekteringssystem för musik, vilket avslöjar sårbarheter för plattformsbias och ljudmanipulation; och (3) hur man omformulerar Förklarbar AI (XAI) som ett kommunikationsproblem, vilket identifierar risker för feljusteringar mellan AI-förklaringar och mänsklig tolkning. Resultaten bidrar med praktiska metoder för musikanalys och detektering med hjälp av AI, och belyser viktiga begränsningar i nuvarande tillvägagångssätt. Sammantaget lägger denna avhandling en grund för framtida arbete med robusta utvärderingsramverk och förbättrade förklaringsmetoder skräddarsydda för AI-system för musik.

Place, publisher, year, edition, pages
Stockholm, Sweden: KTH Royal Institute of Technology, 2025. p. xxi, 83
Series
TRITA-EECS-AVL ; 2025:104
Keywords
Symbolic music representation, AI music detection, AI music, Generative AI, Explainable AI, Symbolisk musikrepresentation, AI-musikdetektering, AI-musik, generativ AI, förklarbar AI
National Category
Musicology Computer and Information Sciences Computer Sciences Music
Research subject
Speech and Music Communication
Identifiers
urn:nbn:se:kth:diva-372384 (URN)978-91-8106-460-5 (ISBN)
Public defence
2025-12-08, Q2, Malvinas väg 10, Stockholm, 15:00 (English)
Opponent
Supervisors
Projects
MUSAiC
Note

QC 20251106

Available from: 2025-11-06 Created: 2025-11-05 Last updated: 2025-11-19Bibliographically approved

Open Access in DiVA

fulltext(863 kB)117 downloads
File information
File name FULLTEXT01.pdfFile size 863 kBChecksum SHA-512
8f05fc5f2550094e81a0e0fbf2ea75d0d3d308b05b077c07bc7e20023dab515cb1fbc45823764268f5bd19169634e99b0ec0ab267465e05816580cdf5b2ea1ee
Type fulltextMimetype application/pdf

Other links

Publisher's full text

Authority records

Cros Vila, LauraSturm, Bob

Search in DiVA

By author/editor
Cros Vila, LauraSturm, Bob
By organisation
Speech, Music and Hearing, TMH
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 118 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

doi
urn-nbn

Altmetric score

doi
urn-nbn
Total: 1438 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf