Sabitlenmiş Tweet
[Life update]: I am starting my PhD @TUDarmstadt. I will be working under the supervision of Prof. @GeorgiaChal on planning and understanding in robotics. Extremely grateful to everyone who made this possible😃!!!
English
Sohan Rudra
416 posts
Sohan Rudra
@rudrasohan
Figuring out life from Intuitions to Equations PhD @CS_TUDarmstadt || Researcher @googleAI 21-23 || Mathematics & Computing @IITKgp 21
Germany Katılım Mayıs 2013
761 Takip Edilen827 Takipçiler
Sohan Rudra retweetledi
Our paper IMPACT was accepted to #RSS2026 🎉
Contact-implicit TrajOpt (CITO) is beautiful… until you try solving it 😅
It lets robots discover contacts instead of hand-designing contact modes.
But in practice?
CITO is hard to solve — and painfully slow.
So we built IMPACT!
English
Sohan Rudra retweetledi
How long should it take to go from 🗣️ "Evaluate π0 on LIBERO" to a real-robot rollout?
We're betting on ⏱️ 1.5 hours.
Introducing 🐚 Nautilus — one prompt. Any policy. Any benchmark. Any robot.
An agentic harness for robot learning + @Anthropic Claude Code plugin.
🔗 lnkd.in/ex3a_QUc
#Robotics #RobotLearning #LLMAgents #ClaudeCode #Harness
English
Sohan Rudra retweetledi
🚀 Hiring PhDs & Postdocs in Structured Robot Learning & Embodied AI @TUDarmstadt (PEARL Lab)
🤖 We study how structure in the robot–environment system can be exploited to learn robust, adaptive, and generalizable behaviors, beyond black-box policies
🔬 Topics:
• Grounding (language → perception → action)
• Structured world models
• VLA + control + memory
• RL (credit assignment, offline→online)
• Whole-body & bimanual mobile manipulation
🇪🇺 New EU Lighthouse Project on Generative AI for Robotics + ERC StG SIREN
👉 Full call: tinyurl.com/pearlgenai
Please repost 🙏
#RobotLearning #EmbodiedAI #Robotics #MachineLearning #PhDPositions #Postdoc
English
Sohan Rudra retweetledi
Excited to be in Korea almost a year after being away from conferences. Now I come back for two conferences #CoRL2025 and #Humanoids2025 with a small guest 👶 While on maternity leave my amazing team continued progress on Structured Robot Learning! Here is a list of the activities of PEARL group at @corl_conf:
1. Today, Saturday Sept. 27 I will give a talk at the corl25-genpriors.github.io talking about Beyond Scale: Structured Priors for Efficient Robot Learning at 11:25. Join me to get a glimpse at the newest cool research from the pearl-lab.com
2. I’ll also be at the panel of the same workshop at 15:25! Looking forward to the fruitful discussions.
3. Also today,Vignesh @FatAndFurious42 will present our new work on extracting Bimanual Affordances from Human Videos (sites.google.com/view/2handedaf…) at the sites.google.com/view/h2r-corl2… and sites.google.com/stanford.edu/c…
4. While Frankie @FrankieZhang41 will present our new work on “Adaptive Diffusion Constrained Sampling for Bimanual Robot Manipulation” (…fusion-constrained-sampling.github.io) at the rational-robots.github.io
5. Tomorrow Sept 28, Steven @softraeh will present SYMDEX: Morphologically Symmetric Reinforcement Learning for Ambidextrous Bimanual Manipulation supersglzc.github.io/projects/symde…. Catch him at Spotlight 2 & Poster session 1.
We have more to present at the Humanoids conference, so stay tuned!
English
Sohan Rudra retweetledi
Such a wonderful first day at #CoRL2025 catching up with old friends and making new ones.
If you’re interested in our work at PEARL and share our vision of structured robot learning for achieving efficient and robust robot behaviors among humans, reach out!
We’re hiring
•PhD positions (through ELLIS: ellis.eu/news/ellis-phd…)
•Postdoc positions (open call: pearl-lab.com/open-positions/)
#RobotLearning #EmbodiedAI #RepresentationLearning #ImitationLearning #ReinforcementLearning #MobileManipulation #DexterousManipulation #HumanRobotInteraction #PhDPositions #PostdocOpportunity
Georgia Chalvatzaki@GeorgiaChal
Excited to be in Korea almost a year after being away from conferences. Now I come back for two conferences #CoRL2025 and #Humanoids2025 with a small guest 👶 While on maternity leave my amazing team continued progress on Structured Robot Learning! Here is a list of the activities of PEARL group at @corl_conf: 1. Today, Saturday Sept. 27 I will give a talk at the corl25-genpriors.github.io talking about Beyond Scale: Structured Priors for Efficient Robot Learning at 11:25. Join me to get a glimpse at the newest cool research from the pearl-lab.com 2. I’ll also be at the panel of the same workshop at 15:25! Looking forward to the fruitful discussions. 3. Also today,Vignesh @FatAndFurious42 will present our new work on extracting Bimanual Affordances from Human Videos (sites.google.com/view/2handedaf…) at the sites.google.com/view/h2r-corl2… and sites.google.com/stanford.edu/c… 4. While Frankie @FrankieZhang41 will present our new work on “Adaptive Diffusion Constrained Sampling for Bimanual Robot Manipulation” (…fusion-constrained-sampling.github.io) at the rational-robots.github.io 5. Tomorrow Sept 28, Steven @softraeh will present SYMDEX: Morphologically Symmetric Reinforcement Learning for Ambidextrous Bimanual Manipulation supersglzc.github.io/projects/symde…. Catch him at Spotlight 2 & Poster session 1. We have more to present at the Humanoids conference, so stay tuned!
English
Sohan Rudra retweetledi
What if your robot's policy is trained on multiple sensors, and one of them becomes unreliable? 🤖📷🤷♀️
Check out our work 'DisDP: Robust Imitation Learning via Disentangled Diffusion Policies', which I will be present at @RL_Conference in RLHF, Imitation learning track.
English
I am deeply honored and grateful to receive the Alfried Krupp Prize, and incredibly proud to bring this prestigious recognition to Technische Universität Darmstadt for the first time.
This award is not only a personal milestone but also a meaningful endorsement of the research we are pursuing in robot perception and learning, toward more structured, intelligent, and embodied robot assistants.
It comes at a pivotal time, as we launch our ERC Starting Grant SIREN research, and it offers crucial support to advance our vision: enabling robots to perceive, reason, and act in complex human environments.
A heartfelt thank you to the Krupp Foundation for this generous support, to TU Darmstadt and hessian.AI, and to my mentors, collaborators, and brilliant team at PEARL who make this journey so rewarding.
I’m especially excited to use this opportunity to grow our research team, and to continue pushing the boundaries of embodied AI together.
linkedin.com/posts/krupp-st…
English
forbesindia.com/lists/ai-list-…
Honored to receive this recognition on behalf of my wonderful team!
English
I'm pleased to announce that, starting October 1st, I will be joining the Computer Science department at Lund University (in Sweden) as a Senior Lecturer.
I will join the Robotic and Semantic Systems group and collaborate in the RobotLab LTH, where I'll bring my RL expertise.
English
Thrilled to share that our paper “CiteFix: Enhancing RAG Accuracy Through Post-Processing Citation Correction” is accepted at ACL 2025 (Industry Track)
We tackle citation errors in RAG systems with efficient, low-latency correction methods.
Read here-: arxiv.org/abs/2504.15629
English
1/ 🚀 Excited to share the first paper from my PhD! We study how to optimally allocate test-time compute between solving problems and verifying them to get the best reasoning performance with your LLM.
(Thread below 👇)
Hritik Bansal@hbXNov
📢Scaling test-time compute via generative verification (GenRM) is an emerging paradigm and shown to be more efficient than self-consistency (SC) for reasoning. But, such claims are misleading☠️ Our compute-matched analysis shows that SC outperforms GenRM across most budgets! 🧵
English
Excited to share my first PhD paper acceptance! 🤓
Our survey on automatic hint generation towards personalized learning was accepted in TACL'25. We highlight the gaps and future possibilities in this exciting interdisciplinary space.
Pre-print: arxiv.org/abs/2404.04728
English
Sohan Rudra retweetledi
At PEARL Lab, we’re advancing robotics to integrate them into human environments. This Christmas, our robots 🤖 joined the festive spirit and made us cookies! 🍪🎄
Wishing you a joyful holiday season from all of us at PEARL Lab!
#Robotics #AI #MobileManipulation #HappyHolidays
English
I'll be joining Nvidia's Applied Deep Learning Research team (nv-adlr.github.io) from May of next year. Excited to kickstart my industry career with the immensely cracked systems team that works on Megatron-LM!
English
Sohan Rudra retweetledi
Honored for the early career keynote at #CoRL2024 last week—unforgettable milestone! Grateful to the community for the recognition. From a small lab in Greece to leading my own group @TUDarmstadt, this journey fuels me to push #RobotLearning further. #ERCStG SIREN is the start!🦾
English
English
Thank you so much @emnlpmeeting for this wonderful recognition! I’m so honored and humbled 💕 Thanks @gneubig for your support throughout!
We’ve been working on this for 1.5 years and everyone who has spoken with me in the recent past knows how passionately I feel about this work and this task in general 🥹 Thanks again to everyone who recognized and appreciated our efforts ✨
EMNLP 2026@emnlpmeeting
#EMNLP2024 Best Paper 1/5: An image speaks a thousand words, but can everyone listen? On image transcreation for cultural relevance
English
Sohan Rudra retweetledi
How much structure does Robot Learning need?
Approaches presented by @GeorgiaChal combine structured robotics priors with learned components
Exciting new ideas in a fresh first keynote at @corl_conf!
English
@sanad_maker @EdinburghNLP A very hearty congratulations, Sanad 🥳
English
I’m thrilled to share that I’ve graduated with Distinction from the M.S. Speech & Language Processing (NLP & AI) program at The University of Edinburgh @EdinburghNLP (1/n)
English
@shahdhruv_ @Princeton A very hearty congratulations, Dhruv 🎉🎉🎉
English
Excited to share that I will be joining @Princeton as an Assistant Professor in ECE & Robotics next academic year! 🐯🤖
robo.princeton.edu
I am recruiting PhD students for the upcoming admissions cycle. If you are interested in working with me, please consider applying.
English