# News · SEPT Lab at USC

Canonical page: https://septlab.com/news.html

## 2026

### June 2026MC\-ORAM appears at ISCA 2026

[MC\-ORAM: A Mask\-Assisted and Counter\-Based Non\-Deterministic ORAM Inside VM\-Based TEEs](https://septlab.com/publications.html#mc-oram-isca26), joint work with Yongqin Wang, Rachit Rajat, Jonghyun Lee, and Murali Annavaram, appears at ISCA 2026\.

### May 2026Hollow\-LLM Attack appears at IEEE S&P 2026

[Hollow\-LLM Attack: Computationally Trivial Weights in Zero\-Knowledge Verification of LLM Inference](https://septlab.com/publications.html#hollow-llm-sp26), by Chen Gong, Beijie Liu, and Mengyuan Li, appears at the 2026 IEEE Symposium on Security and Privacy\.

### April 3, 2026Talk at the University of Chicago

Prof\. Li gave a talk titled *Confidential Computing and Trusted Execution Environments: Challenges and Opportunities for Trustworthy AI* at the [University of Chicago Department of Computer Science](https://computerscience.uchicago.edu/events/event/mengyuan-li-usc-confidential-computing-and-trusted-execution-environments-challenges-and-opportunities-for-trustworthy-ai/)\.

### March 2026WAVE appears at ASPLOS 2026

[WAVE: Leveraging Architecture Observation for Privacy\-Preserving Model Oversight](https://septlab.com/publications.html#wave-asplos26), by Haoxuan Xu, Chen Gong, Beijie Liu, Haizhong Zheng, Beidi Chen, and Mengyuan Li, is published in the ASPLOS 2026 proceedings\. The artifact is available on [GitHub](https://github.com/sept-usc/Wave)\.

### February 3, 2026SCALE presented at HPCA 2026

[SCALE](https://septlab.com/publications.html#scale-hpca26), which tackles communication bottlenecks in confidential multi\-GPU machine learning, was presented at HPCA 2026 in the Distributed and Multi\-GPU Training session\.

## 2025

### December 2025Writing in Symbiosis at the NeurIPS 2025 Creative AI Track

[Writing in Symbiosis: Mapping Human Creative Agency in the AI Era](https://septlab.com/publications.html#writing-symbiosis-neurips25), by Vivan Doshi and Mengyuan Li, appears in the NeurIPS 2025 Creative AI Track\.

### October 2025RMP entry caching study at HASP 2025

[A Close Look at RMP Entry Caching and Its Security Implications in SEV\-SNP](https://septlab.com/publications.html#rmp-caching-hasp25) appeared at HASP 2025\. AMD published security bulletin [AMD\-SB\-3036](https://www.amd.com/en/resources/product-security/bulletin/amd-sb-3036.html) on October 20, 2025, crediting the researchers\.

### October 2025Chekhov’s Gun appears at ACM CCS 2025

[Chekhov’s Gun: Uncovering Hidden Risks in macOS Application\-Sandboxed PID\-Domain Services](https://septlab.com/publications.html#chekhovs-gun-ccs25), led by Minghao Lin, is published in the ACM CCS 2025 proceedings\.

### September 2025Few\-shot graph OOD detection at ECML PKDD 2025

[Few\-Shot Graph Out\-of\-Distribution Detection with LLMs](https://septlab.com/publications.html#fewshot-graph-ood-ecmlpkdd25), first\-authored by PhD student Haoyan Xu, is published in the ECML PKDD 2025 proceedings\.

## 2024

### July 2024SoK on TEE design choices at ACM ASIACCS 2024

[SoK: Understanding Design Choices and Pitfalls of Trusted Execution Environments](https://septlab.com/publications.html#sok-tee-asiaccs24) appears at ACM ASIACCS 2024\.

### 2024SEPT Lab starts at USC

Mengyuan Li joined the [Thomas Lord Department of Computer Science](https://www.cs.usc.edu/) at USC as an Assistant Professor and started SEPT Lab\.