Blog

From Taming to Thriving: Full Fine-Tuning and Logical Sharpening of Polynomial LLMs

Toward Efficient Private Inference: Evolution of Polynomial Architectures for Llama-2 under MPC In practical deployments of Multi-Party Computation (MPC), nonlinear activation functions in Transfo...

Jianjun Lang January 22, 2026 MPC LLM Systems Llama-2 Polynomial Activation Private Inference

Near-Lossless Activation Function Replacement in ViT for Accelerating MPC Computation

Near-Lossless Activation Function Replacement in ViT for Accelerating MPC Computation Author: Jianjun Lang Publication Date: January 16, 2026 Tags: MPC, Vision Transformer, Knowledge Distillation,...

Jianjun Lang January 16, 2026 MPC Vision Transformer Knowledge Distillation System Optimization

Introducing the Model Zoo Paradigm: From Rapid Prototyping to Systematic Asset Preservation

Introducing the Model Zoo Paradigm: From Rapid Prototyping to Systematic Asset Preservation During the early exploratory phase of the ViT polynomial replacement project, my primary engineering obj...

Jianjun Lang January 15, 2026 Engineering Model Management Vision Transformer MPC Software Architecture

From Script Prototyping to Engineering Architecture: A Refactoring Retrospective on Llama 2 and ViT

From Script Prototyping to Engineering Architecture: A Refactoring Retrospective on Llama 2 and ViT In deep learning research, the evolution of a codebase often follows a process of entropy reduct...

Jianjun Lang January 14, 2026 Engineering LLM Systems MPC Llama Vision Transformer Software Architecture

Taming Llama: How I Conquered Modern LLM Architectures (SwiGLU) for Private Computation

Taming Llama: How I Conquered Modern LLM Architectures (SwiGLU) for Private Computation Abstract As large language model (LLM) architectures continue to evolve, activation functions have undergone...

Jianjun Lang January 12, 2026 MPC Private Inference Large Language Models SwiGLU Polynomial Activation Quantization Systems for ML

A Deep Retrospective: Refactoring a Matter Protocol Audit System from Regex Matching to Neuro-Symbolic Analysis

Introduction Validating the security consistency of the Matter protocol across its large and heterogeneous multi-language ecosystem—from the rigorously engineered official C++ reference implementa...

Jianjun Lang December 29, 2025 Static Analysis LLM LLM Agent Matter Protocol Security Neuro-Symbolic

From Spec to Code: Closing the Last Mile in Protocol Security Analysis

Introduction Since September 2025 I have been working with Dr. Dongfang Zhao on ProtocolAnalysisPipeline, an automation effort for protocol security. The idea is simple to state and hard to execut...

Jianjun Lang November 9, 2025 Protocol Security LLM LLM Workflow Systems Security

Rust Verification on AWS: My Journey in Formal Verification during Practicum

Introduction In the Fall of 2024, I had the opportunity to participate in a practicum project at Carnegie Mellon University in collaboration with Amazon Web Services (AWS). The project focused on ...

Jianjun Lang December 10, 2024 Formal Verification AWS Rust

Guiding Students Through Deep Learning Assignments

Introduction In the Fall of 2024, I served as a Teaching Assistant (TA) for Carnegie Mellon University’s 14-757 course, AI Applications in Information Security. This course explores the applicatio...

Jianjun Lang December 6, 2024 Deep Learning Neural Networks Teaching Experience Machine Learning

Reflections on My Experience as a Teaching Assistant in AI Applications in Information Security

Introduction In the Fall 2024 semester, I had the opportunity to serve as a Teaching Assistant (TA) for two advanced courses: “AI Applications in Information Security” and “Machine Learning” at Ca...

Jianjun Lang November 26, 2024 AI Machine Learning

Building a Solid Foundation in Cloud Infrastructure and Big Data

Introduction Throughout my Cloud Infrastructure course at Carnegie Mellon University, I experienced a transformative journey into the world of cloud technologies and big data. This course equipped...

Jianjun Lang April 20, 2024 Cloud Computing Big Data Kubernetes Docker Microservices