
Assistant Professor @JHUCompSci | PhD @Princeton | Postdoc & MSE @Penn @GRASPlab | BEng @ShanghaiTechUni. I develop robots that learn and teach.
Baltimore, MD
Joined June 2018
- Tweets125
- Following344
- Followers711
- Likes937
Excited to launch the JHU Alliance AI Lab website! 🚀
Our lab works on the algorithmic foundations of learning and interaction for safety-assured and collaborative embodied AI systems.
Check us out → alliance-ai.cs.jhu.edu/
@JHUCompSci @HopkinsDSAI @JohnsHopkinsIAA
Haimin Hu retweeted
How can robots and humans safely learn and grow together? 🤖🤝🧑🦲 Watch @HaiminHu in action teaching Human-Centered Robotics: Models and Algorithms and learn how he answers this key research question in our New Faculty Q&A: cs.jhu.edu/news/new-faculty-…
Haimin Hu retweeted
For Halloween, @geetachandra21 got our Unitree G1 to crawl like a spider!
Give Geeta (co-advised w/ @HaiminHu) a follow if you're interested in more humanoid demos.
We're doubling down on whole-body locomanipulation research at @HopkinsDSAI @HopkinsEngineer @JHUCompSci -- reach out if interested in working with us!
(credit to Addison Johnson, who helped design the mounts for the hands, and filmed this; and to Geeta who took a full-stack approach to ensuring that no humans or humanoids were harmed during the production of this video! and to @jloganolson for last year's humanoid crawling deo)
sciphylab.org/
Haimin Hu retweeted
.@SpiderMan’s been working on something new at @JohnsHopkins . . . 🕷🤖
🎃 Happy Halloween from JHU CS! 🕸
(Thanks to @geetachandra21 for this *~amazing~* video!)
Haimin Hu retweeted
Long-overdue update: Earlier this summer, I started a faculty position @JHUCompSci
We're fortunate to have recruited a cohort of incredibly strong students to come work w/ me and other faculty at Hopkins (CS and beyond)
sciphylab.org/
Haimin Hu retweeted
NERC is coming to beautiful @EPrinceton this Fall!
Please see the CfP for papers and demos, as well as our call for Rising Stars in Robotics. Submissions due August 20 AoE.
nerc2026.github.io
Haimin Hu retweeted
The Department of Computer Science is pleased to welcome 6 new tenure-track faculty to its ranks this academic year, including @mangahomanga, @jmin__cho, @HaiminHu, & @_krishna_murthy! 🥳
Learn more about our new assistant professors here: bit.ly/4yNI4xf
Haimin Hu retweeted
A copilot flies the plane for you. A coach teaches you to fly on your own. What does it take to build an AI coach?
We formalize AI coaching as a non-cooperative game where the coach targets the learner's independent competence, then use model-free RL to train coaching policies.
Haimin Hu retweeted
Presenting our work at ACL next week! Going back to RLHS a year on, I think the paper wasn’t just about fixing RLHF. Deeper claim: alignment has to consider downstream outcomes, not just the immediate interaction. The more capable these models get, the more that matters. [1/4]
Think your RLHF-trained AI is aligned with your goals?
⚠️ We found that RLHF can induce significant misalignment when humans provide feedback by predicting future outcomes 🤔, creating incentives for LLM deception 😱
Introduce ✨RLHS (Hindsight Simulation)✨: By simulating future outcomes of the interaction before providing feedback, we drastically reduce misalignment and hallucinations 🦾
Haimin Hu retweeted
Everyone says the latest AI agents will be "job-ready" soon, especially after the release of Fable 5 this week. But is that really the case?
Over the past many months, my group and collaborators have been building Agents' Last Exam (ALE), a benchmark designed to test exactly that claim on real digital labor-market work.
My group and collaborators previously have created many of the benchmarks the field runs on, including MMLU, MATH, CyberGym, and ExploitGym. Today, I'm excited to share Agents' Last Exam (ALE): a rolling benchmark that measures whether AI agents can actually perform economically valuable work across a broad range of real-world domains.
With ALE, we evaluated Fable 5, GPT-5.5, Composer 2.5, and other frontier agent systems across more than 1,500 expert-sourced tasks spanning 55 occupations.
The result is both impressive and sobering.
Today's agents can solve a meaningful fraction of professional tasks. But when we look at the hardest tasks, the ones requiring sustained reasoning, deep domain expertise, and reliable execution over long horizons, they are still far from human-level performance.
On ALE's hardest tier, every frontier agent we tested, including Fable 5, achieved a 0% success rate.
The age of useful agents is here.
The age of truly job-ready agents is not.
We hope Agents' Last Exam (ALE) will serve as a new guidepost and north star for developing agents capable of reliably performing economically valuable work across a broad range of domains.
🧵
Haimin Hu retweeted
We extended the deadline to *June 22nd*! Submit your coolest and craziest (in-progress or completed) works on generalist robot safety 😎
Workshop co-organized with the great team: @ArpitBahety @kensukenk @imp_aa @RobobertoMM @ianabraha @LihanZha
Excited to announce our #RSS2026 workshop: "Rethinking What It Means to be 'Safe' for Generalist Robots"! 🛡️🤖
Have new work or videos of robot safety failures? Submit by June 12! 👇
“AI agents will outperform humans at almost all jobs by 2026–2027.” - The forecast is everywhere.
So we built the exam to test that claim, on real labor-market aligned work. On the hardest tier, top agents pass 2.6%.
Meet Agents' Last Exam (ALE), a rolling benchmark measuring whether agents can actually do real jobs. 🧵👇
Haimin Hu retweeted
This week's #PaperILike is "Cooperative Inverse Reinforcement Learning" (Hadfield-Menell et al., 2016).
Seminal work. My favorite part is the simple example showing that demonstration-by-expert is suboptimal.
PDF: arxiv.org/abs/1606.03137
Haimin Hu retweeted
📣 Registration is now open for #NESCW 2026, 3rd Northeast Systems & Control Workshop, hosted this year at @Princeton 🎉
→ Sat May 2: all-day event + evening social ←
🗓️Register by April 6!
🌐nescw.org
Haimin Hu retweeted
New Meta Research 🚀
AI agents are powerful, but don’t stay aligned with you over time.
When preferences shift, they don’t adapt. You correct them once…they repeat the mistake. 🤦
Introducing PAHF: continual personalization where agents learn from feedback to stay in sync.
Haimin Hu retweeted
Academic relaxation ladder:
Undergrad: relaxes from homework by doing extracurriculars
PhD student: relaxes from research by doing homework
Professor: relaxes from admin by doing research
Haimin Hu retweeted
The RSS Pioneers program applications just opened up!
RSS Pioneers is a program geared toward PhD students close to graduation and postdocs who are interested in a career in leading research groups! If you're interested in applying check out the website for more information:
sites.google.com/view/rsspio…
Haimin Hu retweeted
Join us in advancing data science and AI research! The Johns Hopkins Data Science and AI Institute Postdoctoral Fellowship Program is now accepting applications for the 2026–2027 academic year. Apply now! Deadline: Jan 23, 2026. Details and apply: apply.interfolio.com/179059
Haimin Hu retweeted
We're hiring postdocs at the JHU data science and AI institute
For robotics and computer vision applicants: we have multiple new faculty like @mangahomanga @anand_bhattad Sarah Chiu @jmin__cho @HaiminHu (in addition to several existing faculty)
Apply here
Join us in advancing data science and AI research! The Johns Hopkins Data Science and AI Institute Postdoctoral Fellowship Program is now accepting applications for the 2026–2027 academic year. Apply now! Deadline: Jan 23, 2026. Details and apply: apply.interfolio.com/179059
Haimin Hu retweeted
2025 AAAI / ACM SIGAI Doctoral Consortium interviews compilation ift.tt/E89Z1oR


























