Authenticity
Detect synthetic and manipulated media beyond superficial artifacts.
IEEE ICASSP 2027 · Special Session
Security and Trustworthiness of AI-Driven Signal and Structured Data Processing
A cross-domain forum for secure, private, and reliable AI-driven processing across signals and structured data.
Venue
IEEE ICASSP 2027
Format
Special Session
Focus
Secure & Trustworthy AI
Organizers
4 organizers
01 / Overview
Trust must begin at the signal.
AI-driven systems increasingly transform speech, audio-visual media, sensor streams, source code, and other structured data into consequential decisions and actions. Their expanding autonomy also enlarges the attack surface: synthetic media undermines authenticity, corrupted observations trigger hallucinations, malicious instructions compromise tool use, hidden triggers alter agent behavior, and identity leakage threatens speaker privacy.
These challenges share a signal-processing foundation. Trustworthy decisions depend on how models represent, align, filter, validate, and act on information under noise, manipulation, distribution shift, and adversarial control.
Why this session
One forum for an increasingly connected threat surface.
This session connects work that no single Technical Committee fully captures, promoting transferable defenses, shared terminology, and reproducible cross-modal evaluation.
Reliable inference
Reduce hallucination under noise, uncertainty, and distribution shift.
Agent security
Protect reasoning, memory, tool selection, and downstream actions.
Privacy
Limit identity and sensitive-information leakage across modalities.
02 / Topics
Topics of interest
We welcome original work that advances secure, private, reliable, and robust AI-driven processing across signals and structured data.
-
01
Synthetic media forensics and content authenticity
-
02
Hallucination detection under corrupted and out-of-distribution signals
-
03
Secure and risk-aware tool use by multimodal foundation-model agents
-
04
Backdoor attacks and defenses affecting reasoning, memory, and actions
-
05
Robust multimodal perception under adversarial manipulation
-
06
Privacy-preserving speech and audio-visual signal processing
-
07
Vulnerability and threat classification in source code and structured data
-
08
Cross-modal evaluation protocols, benchmarks, and calibration
03 / Submission
Submission information
Details will follow the official ICASSP 2027 schedule.
Submission dates, author instructions, review policies, and the final session track name will be published through the official conference channels. Please consult the ICASSP 2027 website for authoritative updates.
Visit the official conference websiteRelevant IEEE Signal Processing Society Technical Committees
- Information Forensics and Security
- Machine Learning for Signal Processing
- Speech and Language Processing
- Audio and Acoustic Signal Processing
- Image, Video, and Multidimensional Signal Processing
- Multimedia Signal Processing
04 / Organizers
Session organizers
An international team spanning trustworthy machine learning, multimodal perception, agent security, signal processing, and computer vision.
Assistant Professor in Machine Learning
Chen Feng
Queen's University Belfast, UK
Chen Feng received his B.Eng. from Nankai University, M.Eng. from Tsinghua University, and Ph.D. from Queen Mary University of London. He previously served as an EU Horizon Junior Fellow visiting the University of Trento, Italy, and as a Leverhulme Researcher at University College London (UCL). His research focuses on trustworthy artificial intelligence and machine learning theory.
Associate Professor
Xinyuan Qian
University of Science and Technology Beijing, China
Xinyuan Qian is an Associate Professor in the School of Computer and Communication Engineering at the University of Science and Technology Beijing. She received her Ph.D. in Computer Science from Queen Mary University of London and was a Ph.D. intern at FBK from 2017 to 2018. Her research spans intelligent speech processing and multimodal audio-visual perception and interaction. She has published over 70 papers in venues including TASLP, TMM, CVPR, and AAAI, and has led or participated in projects funded by the NSFC, Beijing Natural Science Foundation, and National Key R&D Program of China.
Postdoctoral Research Fellow
Xueyi Zhang
National University of Singapore, Singapore
Xueyi Zhang received his B.Eng. degree from Dalian Maritime University and his Ph.D. degree from the National University of Defense Technology. He was selected for the Young Elite Scientists Sponsorship Program by the China Association for Science and Technology. He has served as an Area Chair for ACM Multimedia 2026 and as a Senior Program Committee member for AAAI 2027. He is currently a Postdoctoral Research Fellow at the School of Computing, National University of Singapore. His research interests include AI agent security and multimodal manipulation signal detection.
Principal Scientist & Associate Professor
Georgios Tzimiropoulos
Samsung AI Center, Cambridge & Queen Mary University of London, UK
Georgios Tzimiropoulos received the Diploma degree in electrical and computer engineering from the Aristotle University of Thessaloniki, Greece, and the M.Sc. and Ph.D. degrees in signal processing and computer vision from Imperial College London. He is currently Head of the Future Interaction Programme and Principal Scientist at Samsung AI Center, Cambridge, and an Associate Professor at Queen Mary University of London. Previously, he was an Associate Professor at the University of Nottingham and a Senior Research Fellow with the Intelligent Behaviour Understanding Group (iBUG) at Imperial College London.
Contact