Scholar
Yuxuan Zhu
Google Scholar ID: VxMpvw0AAAAJ
PhD student, University of Illinois Urbana-Champaign
Citations & Impact
Citations
138
H-index
5
i10-index
5
Publications
13
Co-authors
7
list available
Publications
Resume (English only)
Academic Achievements
- Accelerate Aggregation Queries with JOINs over Unstructured Data (In submission)
- Breaking Barriers: Do Reinforcement Post Training Gains Transfer To Unseen Domains? (Preprint)
- Teams of LLM Agents can Exploit Zero-Day Vulnerabilities (Preprint)
- Establishing Best Practices for Building Rigorous Agentic Benchmarks (NeurIPS 2025, first place at Berkeley AgentX Summit (Benchmark & Evaluation Track))
- ELT-Bench: An End-to-End Benchmark for Evaluating AI Agents on ELT Pipelines (VLDB 2026)
- UTBoost: Rigorous Evaluation of Coding Agents on SWE-Bench (ACL main 2025)
- CVE-Bench: A Benchmark for AI Agents' Ability to Exploit Real-World Web Application Vulnerabilities (ICML 2025 Spotlight, SafeBench winner, adopted by US AISI, second place at Berkeley AgentX Summit (AI Safety & Alignment Track))
- PilotDB: Database-Agnostic Online Approximate Query Processing with A Priori Error Guarantees (SIGMOD 2025)
- Efficient Approximate Query Processing with Block Sampling (CIDR 2025)
- FedTrans: Efficient Federated Learning via Multi-Model Transformation (MLSys 2024)
- SlabCity: Whole-Query Optimization using Program Synthesis (VLDB 2023)
- An Energy-efficient Computing Offloading Framework for Blockchain-enabled Video Streaming Systems (GlobeCom 2022)
- Sharding for Blockchain-based Mobile Edge Computing System: A Deep Reinforcement Learning Approach (GlobeCom 2021)
Background
- Research interests: data + AI/ML; Research focus: developing statistically grounded approaches to enable efficient data analytics, rigorous AI evaluations, and AI for safety.
