Jailbreak algorithms have costs that aren't accessible to everyone. That's why we created open-source artifacts to benchmark jailbreak attacks & defenses.

Submit artifacts and join the leaderboard at jailbreakbench.github.io, or access artifacts with `pip install jailbreakbench`!

How to write the Introduction?

As a junior student, writing the introduction of a research paper is arguably the most daunting part of paper writing. 😱

Here is a simple template I find useful:

3 Figures 🖼️ + 5 Questions 🤔

📢 I'm recruiting a postdoc at UMD
🗓️ Starting in 2024
🖥️🙋 Topics related to human-AI interaction
🔜 Apply by May 31, but earlier if possible!
🤑 $75-$80k+benefits, unencumbered (ie not 'big project') funding

Link to full post and form for applications in thread.

>

The UMD security group is hiring a postdoc. Want to work in applied cryptography and zk proofs with me or Gabe Kaptchuk? Or usable security, ai security, network security, pq crypto, or programing languages + security. Come to UMD and live in Washington, DC. obj.umiacs.umd.edu/cybersecurity/…

The Maryland Cybersecurity Center—a unique multidisciplinary research and education powerhouse outside of Washington, D.C.—is seeking applicants for its competitive postdoctoral fellowship program. Applications are due March 28. Learn more: cyber.umd.edu/joinus

Academics: 'You should finish your PhD with three papers that you are decidedly passionate about.'

Job market: 'Minimum requirement: 8 top-tier conference papers for research scientist roles.'

This is precisely what statisticians used to say. 'If you can predict well the next data point, they you understand the reality or the process that generates that point'. They said it and said it until Causal Inference proved them wrong. Understanding takes more than prediction.…

Call for papers on deep learning security and privacy. Please submit! website: dlsp2024.ieee-security.org

Happy to share our recently accepted paper at IEEE S&P (IEEE S&P) 2024. arxiv.org/abs/2312.06227

We already know that AI-powered coding assistant tools (e.g., ChatGPT, Copilot, and IntelliCode) may be vulnerable to poisoning attacks and suggest insecure code for developers.

Launching Lean Co-pilot for LLM-human collaboration to write formal mathematical proofs that are 100% accurate. We use LLMs to suggest proof tactics in Lean and also allow humans to intervene and modify in a seamless manner. github.com/lean-dojo/Lean…

Automating theorem proving…

It’s more important than ever to detect whether text was output from an LLM 🔍. We made Mark My Words, a systematic benchmark to evaluate text watermarking schemes for language models. 🧵1/5

Paper: arxiv.org/pdf/2312.00273…

Github: github.com/wagner-group/M…

How do you conduct a SCIENTIFIC evaluation for fuzzing research?
Blindly run different fuzzers regardless of their settings for 24 hours and compare their raw coverage number?
Here are a few tips I found that fuzzer practitioners and researchers often ignore.

There are a lot of new papers on instruction tuning/RLHF this year. I wrote a blog post to give a brief review.

gaotianyu.xyz/blog/2023/11/3…

Join us for the final Fall 2023 Research Seminar Series, marking the conclusion of this semester's enlightening journey of academic exploration and discovery.

Don't miss our Talk featuring Yizheng Chen.
More info: go.umd.edu/DeptTalk

Sponsored by UMIACS & UMD's Values-Centered AI Initiative, a workshop to discuss the fast progression of LPMs was organized by Hal Daumé III, Navita Goyal, Zijian "Jason" Ding, John Kirchenbauer & Yizheng Chen. The keynote speaker was Jacob Andreas.

Read more: go.umd.edu/49NlyIr

I just wanted to announce that the CS Department at <insert any American University> is hiring <insert number between 5 and 15> tenure track faculty in every field of Computer Science broadly defined.
If you or your friends are professorable, you should definitely apply.

Check out the amazing works of our 2023 'Rising Stars in Machine Learning' cohort:

- Megha Srivastava, Stanford University, web.stanford.edu/~meghas/

- Han Shao Han Shao , Toyota Technological Institute at Chicago: sites.google.com/view/hanshao

- Wanrong Zhu Wanrong Zhu , University…

Come be my colleagues! 🔥🔥🔥

Our department UMD Department of Computer Science is hiring multiple faculty! 🚀
We are particularly interested in

Artificial Intelligence, Machine Learning 🤖🧠
Quantum Computing ⚛🌌

but, exceptional candidates in all areas are considered!
cs.umd.edu/job/2023/12895…

Claims of lower standards for women due to diversity initiatives often stem from misconceptions. While the reality can be complex and varied across fields, a nuanced, data-driven approach is crucial to understand and address systemic challenges. #Diversity #Inclusion #Equity

🌟 Calling all aspiring Machine Learning stars! 🌟

One week left to apply to the 'Rising Stars in machine Learning' program at Univ. of Maryland! 🚀

📅 Save the date: Nov 14-15
🏛️ Location: Brendan Iribe Center
👩‍🔬 Present your research
🤝 Network with ML pros
🏨 Travel & lodging…