Research Interest

My research centers on trustworthy AI, with interpretability at its core. I design models and tools that help users understand how AI systems reach their decisions. I also study related challenges in security, fairness, and agent design, aiming to keep AI agents reliable and protected against misuse. Ultimately, I seek to create AI systems that work well in practice while remaining transparent, fair, and resilient.

Education

• Ph.D. in Computer Science, George Mason University, Fairfax, VA (Expected 2028)
• Master of Computer Science, University of Virginia (2021–2023)
• Bachelor of Computer Science, Xidian University (2016–2021)

Publications

• [ACL 2025 Main] Wei B*, Zhu Z. ProtoLens: Advancing Prototype Learning for Fine-Grained Interpretability in Text Classification.
• [In submission] Wei B*, Zhu Z. Neural Symbolic Logical Rule Learner for Interpretable Learning.
• [In submission] Wei B*, Zhu Z. Learning to Explain: Prototype-Based Surrogate Models for LLM Classification.
• [In submission] Raj C, Wei B*, Zhu Z. VIGNETTE: Socially Grounded Bias Evaluation for Vision-Language Models.
• [In submission] Fazli M, Wei B*, Zhu Z. Mitigating Hallucination in Large Vision-Language Models via Adaptive Attention Calibration.
• [MSc thesis] Wei B*, Jia Y, Wang H. An Empirical Study of Neural Contextual Bandit Algorithms.

Internship

AI Agents Developer, Fluency Security
Jun. 2025 – Present

• Designed and implemented a multi-agent system for security ticket analysis using the MCP framework
• Centralized agent initialization via a configurable file enabling users to customize prompt settings without modifying the codebase
• Integrated the client-side application with a RESTful backend service, establishing a robust communication pipeline for real-time data streaming

GenAI Engineer, GoEngage
Jun. 2025 – Present

• Implemented a semantic search engine that replaced brittle keyword matching and improved retrieval accuracy
• Developed an LLM-powered agentic chatbot that autonomously queries backend APIs and generates clear analytical reports for non-technical users

Awards & Achievements

• Scholarship for Academic Excellence of Xidian University (top 3%)
• Outstanding Student of the School of Computer Science and Technology (top 3%)

Professional Service

• Reviewer / Sub-reviewer: ARR (Dec 2024; Feb 2025—ACL; May 2025—EMNLP), KDD 2024, ACML 2024–2025, SSCI 2025, CAIS.