Song Liao
I am a Tenure-Track Assistant Professor in the Department of Computer Science at Texas Tech University. My research interests include IoT security/privacy and data science. Specifically, my work is focused on exploring the security and privacy of Voice Personal Assistants (VPA), such as Amazon Alexa and Google Assistant.
I earned my Ph.D. from Clemson University, where I had the privilege of working under the guidance of Dr. Long Cheng. Before pursuing my Ph.D., I completed both my bachelor’s and master’s degrees at Xi’an Jiaotong University, where I conducted research at the Institute of Artificial Intelligence and Robotics (IAIR) under the mentorship of Dr. Yuehu Liu.
Research Areas
- Policy and Privacy Compliance
- IoT Security and Privacy
- LLM Agent Security
I am actively seeking motivated Ph.D. students interested in the security, privacy, and compliance of LLM agents.
News
-
[Grant] 08/2026: The proposal "CyberAI Innovation: Securing and Leveraging Agentic AI for Cyber Defense" was awarded by the NSF CyberAI program. Thanks, NSF! [Paper] 08/2026: One paper on Reliability-Aware AI Agent Tool Retrieval via Execution Dynamics was accepted to CIKM 2026. Congrats to Weiliang! [Paper] 06/2026: One paper on Diffusion-Aware Error Correction Code Watermarking Framework was accepted to SMC 2026. Congrats to Huan! -
[Service] 05/2026: I will serve as a TPC member for ACSAC 2026, NDSS 2027, and USENIX Security 2027. [Paper] 01/2026: One paper on Adversarial Attacks on Differentially Private Factorization Machines in Recommendation Systems was accepted to SACMAT 2026. Congrats to Felix! -
[People] 08/2025: Congrats to Huan on receiving the Project ENRICH Scholarship from Texas Tech University! -
[People] 08/2025: Weiliang and Huan joined our group. Welcome! [Paper] 06/2025: One paper on Google Play App Account Deletion was accepted to USENIX Security 2025. -
[Service] 04/2025: I will serve as a TPC member for ACSAC 2025, NDSS 2026, and WWW 2026. [Paper] 09/2024: One paper on Acoustic Privacy Notice for Voice Applications was accepted to NDSS 2025. [Paper] 07/2024: One paper on understanding Security and Privacy Risks in the RapidAPI Ecosystem was accepted to CCS 2024. [Paper] 07/2024: One paper on Voice Assistant Application User Review Analysis was accepted to UbiComp 2024. [Paper] 03/2024: One paper on IoT Voice Assistant Applications Analysis was accepted to AsiaCCS 2024. [Paper] 01/2024: One paper on Understanding GDPR Non-Compliance in Privacy Policies of Voice Assistant Applications was accepted to WWW 2024. [Paper] 05/2023: One paper on Static Code Analysis of Voice Assistant Applications was accepted to CCS 2023. [Paper] 05/2023: One paper on Tweet Data Analysis was accepted to KDD 2023. [Paper] 08/2021: One paper on Policy-Violation Detection of Voice Assistant Applications was accepted to USENIX Security 2022. -
[Talk] 07/2021: We presented our work on Policy-Violation Detection of Voice Assistant Applications at the FTC's PrivacyCon 2021. [Paper] 05/2021: One paper on Privacy Policy Analysis of Voice Assistant Applications was accepted for publication in IEEE Security & Privacy Magazine (invited). -
[Award] 12/2020: Our paper received Distinguished Paper Award at the ACSAC 2020. -
[Award] 12/2020: Google awarded us an Abuse Research Grant to support our research on abuse risks in Google products. [Paper] 09/2020: One paper on Trustworthiness of Skill Certification in Voice Assistant Platforms was accepted to CCS 2020. [Paper] 08/2020: One paper on Privacy Policy Analysis of Voice Assistant Applications was accepted to ACSAC 2020. -
[Talk] 08/2020: We presented our work on the Trustworthiness of Skill Certification in Voice Personal Assistant Platforms at the FTC's PrivacyCon 2020. -
[Media] 07/2020: The Register and some others reported our work on measuring the ineffectiveness of privacy policies for voice apps. -
[Award] 07/2020: Google awarded us a $5,000 bug bounty for discovering vulnerabilities of Google Actions. -
[Media] 07/2020: The Register, Washington Internet Daily, ZDNet and some others reported our work on measuring the trustworthiness of Alexa's skills certification.
