Learning Agents Research Group

Learning Agents Research Group, Department of Computer Science, The University of Texas at Austin

[Caroline Wang @CarolineWang98]

[1/n] Just wrapped up 7 months interning with @pcastr at DeepMind and I'm so excited to share our work: https://t.co/SsHsksxO3i.
TLDR: We used LLM-powered program synthesis to automatically model and discover differences between human and LLM strategic behavior

[Peter Stone @PeterStone_TX]

Proud of our latest Nature publication, led by the Sony AI ethics team!

[Peter Stone @PeterStone_TX]

After 6 years as director of Texas Robotics, I'm delighted to hand over the reigns to the capable hands of Prof. Jose Millan. I'm honored to continue supporting Jose in my new role of "Founding Director" as I turn my main focus to acting as Chair of the UT Computer Science Dept

[Brett Barkley @bebark99]

(1/n)
With over 1,300 citations, MBPO is often cited as proof that model based RL beats model free methods. In https://t.co/xq3WXslh67 we showed it often completely fails in DeepMind Control. In our new work, Fixing That Free Lunch (FTFL), we explain why and make it succeed.

[Peter Stone @PeterStone_TX]

UTCS is hiring for several positions this year. Please share with anyone who may be interested!
https://t.co/HSVCf3goLw

[Youngju Yoo @yyoungju_yoo]

Humans learn new manipulation skills from examples and improve as they see more examples. How can we endow robots with the same ability? 🤖

🚀We introduce RoboSSM, scalable in-context imitation learning that enables robots to learn and improve at test time—robots can improve with more examples without returning to the GPU for fine-tuning.
🧵
Paper: https://t.co/cnB6ZVuLEL
Video: https://t.co/kJBYALSQPV

[Caroline Wang @CarolineWang98]

[1/4] 🚀 We’re excited to announce the v1 release of JaxAHT – a new library for Ad Hoc Teamwork (AHT) research, built with JAX for speed & scalability! Check it out 👉 https://t.co/Vmpbm72YwS #AI #MARL #ReinforcementLearning #JAX #AdHocTeamwork

[Michael Munje @michaelmunje]

[1/8] New social navigation paper + benchmark: SocialNav-SUB 🚶🤖 Recent work puts VLMs on robots for navigation, but can they really interpret scenes and extract key details for social navigation? 🔎 https://t.co/2rlcIQpf6h

[Peter Stone @PeterStone_TX]

Congrats to Jiaxun Cui, the 29th Ph.D. graduate from my lab, on the defense of her dissertation entitled "Communication and Generalization in Multi-Agent Learning". Pictured here with me and the rest of her thesis committee (Amy Zhang, Yuke Zhu, Sandeep Chinchali, Yuandong Tian)

[Cameron Angliss @cameronangliss]

Introducing VGC-Bench, a new plug-and-play benchmark made to support research efforts in the domain of Pokémon VGC! Read on for the open source code, the paper, and everything you need to dive in.
[1/9]

[Peter Stone @PeterStone_TX]

If the communication among autonomous vehicles is done in natural language, in principle, human drivers could communicate with them as well. This paper lays the groundwork for that happening!

[Jiaxun Cui 🐿️ @cuijiaxun]

[1/n] Can LLM Agents learn to communicate and coordinate in natural language in driving scenarios 🚗 through self-play interactions? Our recent research shows the potential for such learning; interestingly, there is evidence that they develop concise protocols for collaboration! A distilled version of the agents’ model could generalize to multiple scenarios, communicate at 250 bytes per message, and make decisions within 500 ms while maintaining the original (large) model’s performance.
💬Natural Language Communication among Autonomous Agents
✨Multi-agent Gymnasium for Policy Learning
👀Partial Observation and Negotiation Tasks
More videos & analysis
project page: https://t.co/KYrgvj6Waj
arXiv: https://t.co/Mq608FbslP
with @ChenTangMark, Jarrett Holtz, Janice Nguyen, @aleallievi, @HangQiu, @PeterStone_TX

open to discussion & collaboration!
@utlarg @texas_robotics
#LLM #multiagent

[Jiaheng Hu @JiahengHu1]

Real-world RL, where robots learn directly from physical interactions, is extremely challenging — especially for high-DoF systems like mobile manipulators.
1⃣ Long-horizon tasks and large action spaces lead to difficult policy optimization.
2⃣ Real-world exploration with whole-body contact raises serious safety concerns.

🚀 Introducing SLAC, a framework that brings safety and efficiency to whole-body real-world RL.
Paper: https://t.co/TK8ylbAtdL
Video: https://t.co/9YFEL8metM
🧵

[Peter Stone @PeterStone_TX]

And congrats to co-first-author Zhihan as well!

[Learning Agents Research Group @utlarg]

👏

[Harshit Sikchi @harshit_sikchi]

Our work on unsupervised RL, Proto Successor Measures (PSM), will be presented at #ICML2025. PSM learns from offline reward-free data to give near-optimal policies at test time for any reward function with a new representation objective.

[Learning Agents Research Group @utlarg]

Huge congratulations to @PeterStone_TX for receiving the ACM Allen Newell Award! 🎉 We are proud to celebrate this incredible recognition of his groundbreaking contributions to AI and computer science.👏 #ACM #AAAI #AllenNewellAward

[Association for Computing Machinery @TheOfficialACM]

🙌 Meet the 2024 ACM Technical Awards Recipients!
We’re proud to honor this year’s innovators in autonomous systems, cryptography, and software for parallel computers:

🏆 Peter Stone – ACM-AAAI Allen Newell Award
For significant contributions to the theory and practice of artificial intelligence (AI). @UTAustin @SonyAI_global
🔗 https://t.co/U4vpUvBgyX

[Peter Stone @PeterStone_TX]

Congratulations to my Ph.D. student, Jiaheng Hu - co-advised with Roberto Martin-Martin.
https://t.co/NFW0PCBHtu

[Siddhant Agarwal @agsidd10]

Check out our work for further details.
Paper: https://t.co/YoRzN1TEQE
This work couldn’t have been possible without amazing collaborators: @harshit_sikchi @PeterStone_TX and @yayitsamyzhang

@utlarg @lab_midi @texas_robotics @MLFoundations(4/n)

[Zizhao Wang @duke_zzwang]

In multi-object env, why do most Unsupervised Skill Discovery methods fail to learn complex skills like tool use? Because they simply maximize state coverage. Introducing our solution SkiLD: Skill Discovery Guided by Factor Interactions (NeurIPS24) https://t.co/buo3qSdI1O

[Jiaheng Hu @JiahengHu1]

🚀Unsupervised RL can learn skills purely from reward-free interactions with an environment. But what form of skills can facilitate efficient downstream hierarchical learning? Introducing DUSDi: Disentangled Unsupervised Skill Discovery (NeurIPS24). https://t.co/nlf2paQMvE 🧵