Skip to content
Log in
Sponsored AI Networking Cookbook: Practical recipes for AI-assisted network automation and development AI Networking Cookbook: Practical recipes for AI-assisted network automation and development

Author

Wencen Wu

h-index 1 View on Semantic Scholar

We have 2 of 8 papers

We haven’t gathered this author’s papers yet. Follow them and we’ll fetch their work.

Not the right person? Other researchers publish under this name.

Conference Jul 2026

SpartanAI: An AI Agent for Automated Course Review

Quality assurance in online and blended course design remains labor-intensive, costly, and difficult to scale across institutions. Faculty often lack timely access to instructional design support aligned with established quality standards such as the Quality Matters (QM) rubric. While instructional designers bring deep expertise in online course design, faculty often lack equivalent familiarity with evidence-based design principles; SpartanAI provides these instructors with structured, rubricgrounded feedback that would otherwise require specialized expertise or a formal review process to obtain. While recent advances in large language models (LLMs) have enabled automated analysis of textual artifacts, most AI applications in education remain student-facing and do not provide systematic, rubricaligned course review services for instructors. In this paper we present SpartanAI, an LLM-enabled AI Framework that automates QM-aligned course review through structured learning management system (LMS) ingestion, module-level content extraction, semantic indexing, and rubric-by-rubric evaluation. Instructors upload exported course packages which are parsed into structured representations, indexed using vector embeddings, and evaluated rubric by rubric. For each rubric, the system retrieves relevant course artifacts, generates alignment scores, identifies supporting evidence, and provides actionable revision guidance. Initial evaluation against QM-certified reviewer feedback on a pilot set of courses shows promising alignment in rubric-level scoring and feedback relevance. These results suggest that structured LLM-based evaluation pipelines can augment institutional course quality assurance processes and provide scalable, instructor-centered AI support.

Pratik Korat, Darshan Patel, M. Eirinaki et al. · 0 citations
Conference Jul 2026

A Local Middleware for Privacy-Preserving LLM Inference via Reversible Entity Substitution

Large Language Models (LLMs) such as ChatGPT, Gemini, Claude, etc. have become integral tools for technical writing, software development, and communication. However, these systems, despite privacy assurances, still expose a significant privacy risk. User prompts often contain personally identifiable or proprietary information that is transmitted in plain-text to external providers, where it may be logged or accessed despite opt-out policies. This research aims to address this gap by designing a lightweight local encryption middleware that acts as a privacy firewall between users and remote AI APIs. Before a message is sent, the middleware automatically detects sensitive entities (Personally Identifiable Information or PII) and replaces them with typed placeholders such as [PERSON_1] or [ORG_1]. A local mapping (with optional authenticated encryption at rest) links these placeholders to their original values, allowing the system to reconstruct (rehydrate) the final response once the model replies. This design aims to reduce the exposure of sensitive information in human-readable form while maintaining the AI output. We evaluate the system in terms of detection accuracy, reconstruction fidelity, and utility preservation, demonstrating a practical solution for privacy-preserving LLM interaction.

Aditya Vishwakarma, Wencen Wu · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.