Zichuan Li

MAD Bugs: Discovering a 0-Day in Zero Day Our intern used Claude to discover and patch a radare2 0-day on his first day at work. This is madness =) blog.calif.io/p/mad-bugs-dis…

@Ri7erLi and I will present our paper "Les Dissonances: Cross-Tool Harvesting and Polluting in Pool-of-Tools Empowered LLM Agents" at NDSS. Location: Coast Ballroom (Session 3B Time: 14:30–15:50 Come by if you're around, happy to chat! 🙌 Paper: ndss-symposium.org/wp-content/upl…

Claude Code is All You Need When I first joined Anthropic I was surprised to learn that lots of the team used Claude Code as a general agent, not just for code. I’ve since become a convert! I use Claude Code to help me with almost all the work I do now, here’s how:

🚀 Exciting update: I'll be graduating in ~6 months with a PhD in Computer Science (Cybersecurity) from @UCRiverside. I’m actively exploring full-time researcher roles—focused on networking, security, wireless, cryptography, and protocol design/attacks. 🏆 My work has led to: • 2x @PwnieAwards nominations (2023, 2025) • 1x Pwn2Own winner • First-authored publications at top venues (2x @acm_ccs, 2x @BlackHatEvents) • Recent research internship with @PaloAltoNtwks. If your team is hiring—or if you know of opportunities—Find my email address at xinanzhou.com, and referrals are hugely appreciated 🙏 #netsec #security #infosec

🚨 New AG2 Talk Alert! 🚨 Join us on Aug 28, 9 AM PST for "Maris: A Security Controlled Development Paradigm for Multi-Agent Collaboration Systems" by @cuijian0819 Jian Cui from UIUC & Berkeley AgentX competition winner! 🛡️ #AI #Cybersecurity RSVP now: discord.com/events/1153072…

SDSU Fully Funded PhD Openings | Systems Security & Privacy, AI Security & Privacy Position offered through the joint SDSU + UCI program. Enrollment starts in Fall 2026, with applications due by December 1, 2025. Advisor’s website: 1phan.com

holy shii, someone from @pb_ctf x BlueWater used an background AI agent solve a LiveCTF challenge while that player was still working on it. youtu.be/TYn38VfmDRU?t=…

🥳🥳

@ZifengKang Congrats!

I am excited to announce that our paper titled “Follow My Flow: Unveiling Client-Side Prototype Pollution Gadgets from One Million Real-World Websites” won the Distinguished Paper Award on IEEE S&P 2025! Great thanks to Dr. Yinzhi Cao and other colleagues! #ieeesp2025

My 10k-word writeup on exploiting a heap-overflow in Llama.cpp's RPC Server's Tensor-operation to RCE. This by far is one of the most challenging but fun exploitation I've ever researched on. retr0.blog/blog/llama-rpc…

So thrilled to share that we team Hoosiers (Jian @cuijian0819 and I) won the second place in the safety track of the LLM Agent Hackathon! We would greatly appreciate Luyi (@luyixing) and Xiaojing (@xiaojingliao)'s guidance. Check out our project on the website!

Mitigating racial bias from LLMs is a lot easier than removing it from humans! Can’t believe this happened at the best AI conference @NeurIPSConf We have ethical reviews for authors, but missed it for invited speakers? 😡

Spent a great summer and learned a lot from ninja folks! Highly recommended!

🚨 Call for Papers NDSS '25 Workshop on Security and Privacy in Standardized IoT (SDIoTSec) 🔑 IoT standards, Matter, IoT security labels, supply chain, etc. 🔐 Security, Privacy, Governance 📝 8 or 4 pages 📅 Deadline: Dec. 6 ndss-symposium.org/ndss2025/co-lo… @NDSSSymposium @FCC @NIST

@xwang_lk I like this post

I was in my mid 20s, I was a CS PhD student in AI, I wanted to build tech startups, I wanted to be a faculty, I wanted to build my own home, I wanted to travel everywhere, I wanted to be financially free before 40. Now I am in my early 30s, achieved some, chasing the others. Life is long. Just do it, step by step. Dreams will come true.

🦀

Exciting update on Project Zero’s LLM research: googleprojectzero.blogspot.com/2024/10/from-n…

@Eqq111e 赤道企鹅 我的偶像

At night towards the end of the day, getting my new RCE. Almost two months of work. 😚

Are we ready for a world where our data is exposed at a glance? @CaineArdayfio and I offer an answer to protect yourself here: tinyurl.com/meet-ixray

“we find LLM-generated ideas are judged as more novel (p < 0.05) than human expert ideas while being judged slightly weaker on feasibility.” 😧