Optimizing robots at deployment time. Views are my own (non-linear echoes of situated experience) @MIT_CSAIL | @USCViterbi | @NSF GRFP Fellow
Cambridge, MA
Joined July 2019
- Tweets271
- Following659
- Followers598
- Likes4.9K
Nathan Dennler retweeted
So excited for this research! I've always been annoyed at the supremacy of verified rewards, and now we can finally show it.
In this work, we develop ways of measuring: how much of a reward signal is verifiable, how much can we put into words, and how much is shared but tacit, or impossible to articulate?
Nathan Dennler retweeted
If you are a rising PhD student close to graduation working in the space of responsible computing and AI, please consider applying to the the Georgia Tech 2026 Summit on Responsible Computing, AI, and Society (RCAIS) Doctoral Consortium (rcais26.github.io/doctoral-c…). Apps due Nov 1.
Nathan Dennler retweeted
“I left my bag on the table.”
For a robot working alongside people, a sentence like this can reveal something its cameras haven’t seen. Yet most mapping systems build their view of the world from onboard sensors alone.
What if language could be a sensor too?
In Language as a Sensor (CoRL 2026), we turn these descriptions into spatial beliefs that can be combined with vision.
📄 arxiv.org/abs/2606.08666
🌐 aryan-naveen.github.io/langu…
w/ @jasonxyliu @lucacarlone1 @andreea7b
🧵 1/6
Nathan Dennler retweeted
Interested in a PhD in HCI at UT Austin CS? Meet us at #UIST2026!
Join Karthik Mahadevan (@karthikm0), Michelle Lam (@michelle123lam), and me to chat about research and ask questions about PhD applications. Our interests span a range of technical HCI areas (see the flyer).
RSVP: forms.gle/a5oFnbDhV9meJ7Fz5
🗓️: Tuesday, November 3, from 3:30 PM (during the afternoon coffee break)
📍: Renaissance Foyer East, Renaissance Center
Looking forward to meeting you!
I recently had the opportunity to give a Rising Stars talk at @NERC_Robotics on algorithms for deployment-time optimization of assistive robot systems.
One of my favorite things about local conferences is getting to see the breadth of interesting robotics problems people are working on! Many thanks to all the organizers!
Nathan Dennler retweeted
New faculty member @_jessethomason_ is bringing his spotlight on robotics to #GTcomputing.
Thomason is exploring how roboticists can create new ways to communicate with robots using computational representations of language.
Read more: cc.gatech.edu/news/new-assoc…
ALT Jesse Thomason
Nathan Dennler retweeted
I’m recruiting 2~3 funded PhD students for Fall 2027 to join the Symbiotic Interfaces Lab at UT Austin CS (@UTCompSci)!
Check out our research:
symbiotic-interfaces.cs.utex…
Apply by Dec 15, mentioning my name:
cs.utexas.edu/graduate/apply
Reposts appreciated!
(1/2)
Nathan Dennler retweeted
Surprisingly a robot can sometimes do highly-dynamic manipulation by just copying the demonstration 🤠. Without any learning, cracking a whip or lassoing a cleat can be done directly from human motion capture. And amazingly this process can be nearly automatically implemented by GPT-6 Astra
Check out the full blog post with fun visuals: krishnasuresh.org/blog/2026/…
Turn on sound for full effect 🔈
Quick side note: the whips shown in this post are called signal/stock whips. These whips are built to generate the loud cracking noise when the tip breaks the sound barrier and is a popular art form: youtube.com/watch?v=ietBtr7s….
Nathan Dennler retweeted
I am open-sourcing DeckWerk, a multi-platform slide editor with stellar video support, real-time collaboration, and Agent support. I used to be on Linux but had to switch to Mac b/c there was no good enough slide editor - @OmarchyLinux & @dhh inspired me to vibe-code my own, so I'm finally back on Linux!
Install from source via GitHub (installers coming soon!):
github.com/vsitzmann/deckwer…
Nathan Dennler retweeted
Myo Action (our CHI2026 paper) shortens human intention-action gap, using ultra-fast EMG & muscle-stimulation loop. learn more about this project:
paper: dl.acm.org/doi/full/10.1145/…; video: youtu.be/M0EVjipRmPU?si=7Jl3…
Researchers at the @UChicago developed an augmentation technology called Myo Action that reduces the time between thinking and moving.
hackster.io/news/myo-action-…
Nathan Dennler retweeted
Excited to announce our new NSF Science and Technology Center for Human and Robot Co-Adaptation! We'll study how people and robots adapt to each other, so that we can build technology that fits human needs.
nsf.gov/news/nsf-launches-th…
Nathan Dennler retweeted
New histogram and CDF visualizations now available in @pyribs 0.12.0. Go check it out! docs.pyribs.org/en/stable/wh…
Nathan Dennler retweeted
That's a wrap for RSS Pioneers 2026! We sincerely thank our sponsors @nvidia and @RoboticsSciSys, reviewers, speakers, mentors, organizers, and the 2026 RSS Pioneers cohort for making this year's program such a meaningful and inspiring experience.
Nathan Dennler retweeted
Super exciting project led by @keplerccccc @Shuangyu0_0Xie !!
It was incredible to see GaP work for so many different tasks, and I’m really glad I got to contribute to a part of it!
Graph-as-Policy (GaP) is a new variant of Agentic Robotics from @NVIDIA and @UCBerkeley that builds computation graphs (like ROS) to ensure modularity, manage complexity, and facilitate interpretability.
🧵Open code and paper: graph-robots.github.io/gap/
#Robotics #Automation #AgenticAI #EmbodiedAI
Nathan Dennler retweeted
Tell a novice "take the foil off the bowl," and you stop one accident. Teach them metal doesn't belong in microwaves, and they'll skip the metal containers too.
SENSEI fixes the misconceptions behind mistakes, not just the mistakes.
So fun co-leading with @ayano_hiranaka👇
Fix the mind, not just the move 🧠
AI assistance shouldn't just correct immediate behavior today. It should understand why mistake happened and help us do better tomorrow.
Excited to share work with @SophieHsu13 @snikolaidis19 @ebiyik_ @daniel_t_seita accepted to #ICML2026! 🧵1/6
Nathan Dennler retweeted
This is a normal day in our lab, but don't tell our advisor! Simulation provides a way to avoid such expensive behaviors, but the consequences of actions, like fluid damage or breakage, aren't observable in sim, making it hard to train and evaluate robot policies for safety.
🧵👇
Nathan Dennler retweeted
How can we get robot hands to “hear” slip and contact through microphones, and react to them?
We’re excited to share VibeAct, an approach that uses piezoelectric microphones embedded in robot fingertips to estimate contact and slip, then learns reactive policies from this tactile feedback!
vibeact.github.io
Nathan Dennler retweeted
After over 10 years at Stanford, it's time to leave :)
I will be joining UT Austin's CS department as an assistant professor in fall 2027! If you're excited to envision the future of interaction with AI, I'm recruiting PhD students this cycle. Come join me!
Nathan Dennler retweeted
I’m thrilled to share that I’ll be joining UT Austin Computer Science (@UTCompSci ) as an Assistant Professor, starting Jan 2027!
I'm recruiting PhD students for Fall 2027, postdocs, and researchers at all levels:
forms.gle/WMbb7NYts1fX5BYG7
I’ll continue my work (yudai-tanaka.com) in Human-Computer Interaction (HCI) and start the Symbiotic Interfaces Lab, where we’ll explore how computational systems & humans can co-adapt to augment human sensorimotor and cognitive abilities. 🦾🧠
Nathan Dennler retweeted
I’ll be at ICRA 2026 this week!
Feel free to reach out or come say hi - happy to chat about robot learning, multimodal human feedback, reward learning, HRI, or anything else :)
Two posters on language-guided robot learning:
June 3 (Wed), 15:00–16:30 Hall C (1–100): MotIF: Motion Instruction Fine-tuning
motif-1k.github.io/
June 4 (Thu), 15:00–16:30 Hall C (201–300): Masked IRL: LLM-Guided Reward Disambiguation from Demonstrations and Language
mit-clear-lab.github.io/Mask…