6 papers
HBF Sucks! A Full-Stack Characterization of High-Bandwidth Flash for KV-Centric LLM Serving
Zhuoran Li, Zhuohang Bian, Xin Huang +3
A faster storage device should make serving faster. We find the opposite. High-Bandwidth Flash (HBF) stacks NAND behind a wide, package-local interface, promising flash-scale capac…
MacAgentBench: Benchmarking AI Agents on Real-World macOS Desktop
Yikun Fu, Bowen Fu, Zhenyu Wu +10
Computer use agents (CUAs) have advanced rapidly in desktop automation, and a growing number of users deploy CUAs such as OpenClaw on Mac Mini for always-on automation. However, ex…
Continuous Discovery of Vulnerabilities in LLM Serving Systems with Fuzzing
Yunze Zhao, Yibo Zhao, Yuchen Zhang +2
LLM inference and serving systems have become security-critical infrastructure; however, many of their most concerning failures arise from the serving layer rather than from model…
Enabling Performant and Flexible Model-Internal Observability for LLM Inference
Nengneng Yu, Sixian Xiong, Yibo Zhao +2
Today's inference-time workloads increasingly depend on timely access to a model's internal states. We present DMI-Lib, a high-speed deep model inspector that treats internal obser…
OS-Themis: A Scalable Critic Framework for Generalist GUI Rewards
Zehao Li, Zhenyu Wu, Yibo Zhao +11
Reinforcement Learning (RL) has the potential to improve the robustness of GUI agents in stochastic environments, yet training is highly sensitive to the quality of the reward func…
Quant Fever, Reasoning Blackholes, Schrodinger's Compliance, and More: Probing GPT-OSS-20B
Shuyi Lin, Tian Lu, Zikai Wang +3
OpenAI's GPT-OSS family provides open-weight language models with explicit chain-of-thought (CoT) reasoning and a Harmony prompt format. We summarize an extensive security evaluati…