12 papers
MemLeak: Cross-User Semantic Leakage in Multi-Tenant AI Agent Memory
Priyanka Mudgal, Kai Zhao, Guilin Zhang +4
Personal AI agents in enterprise multi-tenant deployments share a common vector store for long-term memory. Shared embedding spaces create a surface for cross-user memory leakage:…
Does the Readout Bypass Leak the Input? A Feature-Visibility Audit of Hybrid Quantum-Classical Models
Guilin Zhang, Kai Zhao, Xiquan Cui +3
Readout-side residual hybrids concatenate raw inputs with measured quantum features. Under single-example gradient sharing, a biased first linear layer admits standard analytic rec…
When Error Mitigation Makes Things Worse: Budget-Aware Evaluation, Extrapolation Failure, and the Calibration Trust Boundary
Guilin Zhang, Kai Zhao, Xiquan Cui +3
Error mitigation is expected to turn noisy quantum measurements into more useful estimates. We show that it can instead amplify error. Under control-offset miscalibration, the eval…
Budget Boundary Effects in Test-Time Mathematical Reasoning
Guilin Zhang, Ziqi Tan, Wulan Guo +5
A cumulative token cap can fall inside a mathematical derivation, forcing a test-time controller to choose between stopping at the cap (strict) and allowing the current attempt to…
Report: Progressive Disclosure of Agent Skills
Guilin Zhang, Kai Zhao, Priyanka Mudgal +4
Users of Workday's deployed LLM-based agents often request features which can be addressed by defining named procedures, also known as skills, in the LLM context, effectively augme…
Winning by Peeking: Unenforced Budgets and Test-Set Selection Inflate Short-Budget AutoML Comparisons
Guilin Zhang, Kai Zhao
Comparisons between AutoML systems at short time budgets -- tens of seconds rather than hours -- are common in tool READMEs and workshop papers, and they are easy to get wrong. We…
Deployment-Time Memorization in Foundation-Model Agents
Lei, Chen, Guilin Zhang +8
Foundation-model agents are increasingly long-lived systems that remember users across interactions, making memorization an explicit deployment-time function rather than solely a p…
EvalStop: Using World Feedback to Detect and Correct Reward Overoptimization in Multi-Tenant RLHF Platforms
Guilin Zhang, Chuanyi Sun, Kai Zhao +3
Cloud LLM fine-tuning platforms increasingly serve RLHF workloads, where a learned reward model is optimized as a proxy for human quality. As Gao et al. (2023) showed, this proxy d…
Xpertbench: Expert Level Tasks with Rubrics-Based Evaluation
Xue Liu, Xin Ma, Yuxin Ma +36
As Large Language Models (LLMs) exhibit plateauing performance on conventional benchmarks, a pivotal challenge persists: evaluating their proficiency in complex, open-ended tasks c…
Adaptive Memory Admission Control for LLM Agents
Guilin Zhang, Wei Jiang, Xiejiashan Wang +5
LLM-based agents increasingly rely on long-term memory to support multi-session reasoning and interaction, yet current systems provide little control over what information is retai…
LLMs as Orchestrators: Constraint-Compliant Multi-Agent Optimization for Recommendation Systems
Guilin Zhang, Kai Zhao, Jeffrey Friedman +1
Recommendation systems must optimize multiple objectives while satisfying hard business constraints such as fairness and coverage. For example, an e-commerce platform may require e…
RobustExplain: Evaluating Robustness of LLM-Based Explanation Agents for Recommendation
Guilin Zhang, Kai Zhao, Jeffrey Friedman +1
Large Language Models (LLMs) are increasingly used to generate natural-language explanations in recommender systems, acting as explanation agents that reason over user behavior his…