goosebumped watching this in live. so happy she's back strong💪💪💪

Today Facebook reminded me of a post from 5 years ago, including this nice photo. I was cuddling my little 3-month-old boy while working on building and testing #AFLNet protocol fuzzer—my academic “child”. You can see the Wireshark screen :) Time flies and both have been growing!

Looking for universal, backward-compatible kernel read and write primitives for both ARM and Intel-based macOS systems? No problem! Check it out at: github.com/wangtielei/POC…. The PoC uses only existing kernel mechanisms and does not require complex memory manipulation techniques.

🇦🇺😤Australia coach: China’s record-breaking victory by a body length in 100m freestyle is “not humanly possible”. 🇨🇳🤷🏻‍♂️IOC spokesman: China is the most tested swimming team here, 600 times since Jan…They are fully tested. No decency, just hatred from a defeated supremacist.

Qinwen Queen of Paris! 👑🥇 Qinwen Zheng becomes the first Chinese player to ever win a singles gold medal at the Olympic Tennis Event! #Paris2024 | #Olympics | #tennis

Our paper on ARVO, a huge (5000+) dataset of reproducible (w/PoC), recompilable vulns in open-source C/C++ projects is now up on arXiv! Awesome work by n132, X3eRo0, Jordi Del Castillo, Haoran Xi, Abdelouahab Benchikh, & profs from NYU/ASU/UNSW arxiv.org/abs/2408.02153

# RLHF is just barely RL Reinforcement Learning from Human Feedback (RLHF) is the third (and last) major stage of training an LLM, after pretraining and supervised finetuning (SFT). My rant on RLHF is that it is just barely RL, in a way that I think is not too widely

fyi: Meta's funding a few new grants for faculty to create evaluation benchmarks (deadline Sept 7). Could be of interest! #NLProc llama.meta.com/llm-evaluation…

That text at the top—excellent attention-grabbing poster #acl2024nlp

The NVIDIA Graduate Fellowship application is now open! Deadline: 09/13 research.nvidia.com/graduate-fello… As a past recipient, I strongly recommend applying! You get the money as well as the guaranteed opportunity for a summer internship!

Some statistics for USENIX Security : - 417 accepted papers - 53% increase in submissions - 356 PC members wrote more than 6,600 reviews - 24 (!) papers directly accepted, more stats in the picture #usesec24

CVSS 9.8 Alert 🚨 Microsoft has issued a critical alert for a critical vulnerability CVE-2024-38063 that allows attackers to execute arbitrary code on Windows 10, 11, & Server systems. How the Vulnerability Works? 1/4 #IPV6

If you're going on the CS faculty job market...

Team Atlanta's first #AIxCC blog post is live! 🚀 ✅ Team announcement ✅ Challenge debriefing ✅ Reflections & suggestions Insights for #AI, #Security, #LLM, and #AIxCC communities. Read now: team-atlanta.github.io/blog/post-atl/​​​​​​​​​​​​​​​​ DARPA ARPA-H Perri Adams Daria Bahrami

Now, reembark on the Journey to the West! #BlackMythWukong is now available on PS5, Steam, Epic Games Store, and WeGame. Confront your destiny on the journey westwards!

NDSS review: "this labeling scheme is a joke", "you’ve presenting results that claim it’s a solved problem. It’s irresponsible", "This isn’t an interesting problem and it’s not a useful result to know" Me: can you calm down... Honestly, such review hurts students a lot.

🚀 More insights on LLM and security! Check out our latest blog on how our Cyber Reasoning System, Atlantis, autonomously discovered and patched a real bug in SQLite3 during the AIxCC competition! Read it here: team-atlanta.github.io/blog/post-asc-… #AIxCC #Cybersecurity #LLM #sqlite