Skip to content
View Koukyosyumei's full-sized avatar
108 followers · 13 following

Achievements

Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: Starstruckx2Achievement: Pull Sharkx3Achievement: YOLO

Achievements

Achievement: Pair Extraordinairex2Achievement: QuickdrawAchievement: Starstruckx2Achievement: Pull Sharkx3Achievement: YOLO

Highlights

  • Pro

Block or report Koukyosyumei

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
Koukyosyumei/README.md

Hideaki Takahashi 🌍

I’m a Ph.D. student in Computer Science at Columbia University working on software security, cryptography, and agentic systems.

Keywords: Software Testing, AI, Fuzzing, Formal Verification, Microkernel, Cryptography, Zero Knowledge Proof, Rust, Lean4

🚀 Featured Projects

Project Description Stars
h5i Next-Gen AI-Aware Git GitHub stars
AIJack Security risk simulator for machine learning (10K+ downloads, referenced in 10+ papers) GitHub stars
zkFuzz ZK circuit fuzzer (IEEE S&P'26, 60+ confirmed bugs) GitHub stars
MyZKP Tutorial e-book to implement zero-knowledge proofs from scratch using Rust and CUDA. GitHub stars

📄 Selected Publications

  • [IEEE S&P '26] 0️⃣: zkFuzz: Foundation and Framework for Effective Fuzzing of Zero-Knowledge Circuits
    • Hideaki Takahashi, Jihwan Kim, Suman Jana, Junfeng Yang | paper | code
  • [AAMAS '24] 🚗: On the Transit Obfuscation Problem
    • Hideaki Takahashi, Alex Fukunaga | paper | code
  • [ICLR '24] 🌳: VFLAIR: Benchmark for Vertical Federated Learning
    • Tianyuan Zou, Zixuan Gu, Yu He, Hideaki Takahashi, Yang Liu, Ya-Qin Zhang | paper | code
  • [CVPR '23] 🎥: Breaching FedMD, Image Recovery via Paired-Logits Inversion Attack
    • Hideaki Takahashi, Jingjing Liu, and Yang Liu | paper | code

🐛 Selected Security Findings

AI Agent / MCP servers

Cryptography / Zero-Knowledge Proof

✍️ Pinned Blog Posts

  • Formal Verification of Blockchain Consensus Algorithms in Lean 4 | blog | code
  • Learning x86–64 Machine Language and Assembly by Implementing a Disassembler | blog | code

Pinned Loading

  1. AIJack AIJack Public

    Security and Privacy Risk Simulator for Machine Learning (arXiv:2312.17667)

    C++ 428 66

  2. h5i-dev/h5i h5i-dev/h5i Public

    Next-Gen AI-Aware Git. Auditable sandbox, model/prompt-aware commits, 95% less token waste, 3.5x richer PR brief, 1.8x faster multi-agent loops.

    Rust 319 16

  3. zkFuzz zkFuzz Public

    ZK Circuit Fuzzer (IEEE S&P'26)

    Rust 44 4

  4. MyZKP MyZKP Public

    Building Zero Knowledge Proof from Scratch in Rust

    Rust 52 2

  5. FLAIR-THU/PairedLogitsInversion FLAIR-THU/PairedLogitsInversion Public

    Implementation of "Breaching FedMD: Image Recovery via Paired-Logits Inversion Attack" (CVPR 2023)

    Python 6

  6. Runwai Runwai Public

    COMS 6998 (Fall 2025): Refinement-typed DSL for certified AIR constraints and lookups

    Lean 6 3