Lihe Li

PhD student @UTCompSci, MS, BS @NJU1902 | Reinforcement Learning (RL), Multi-agent Systems (MAS)

[Jiaxun Cui 🐿️ @cuijiaxun]

Accepted as spotlight at NeurIPS 2026 👀 Nice work for continually improve agent collaborativeness (w/ anyone)!! @CarolineWang98 @ArrasyRahman Ben Nativi Johnny Liu @yoonchangsung @PeterStone_TX

[Lihe Li @lilh76]

@rlebench Really exciting work — RLE-Bench looks like a great step toward evaluating truly general-purpose embodied agents. Looking forward to seeing how far current coding agents can go!

[rlebench @rlebench]

GPT-6 Astra is impressive, but agentic robotics ≠ just control.

Physical agents should control robots, learn new skills, and design their own hardware.

So we introduce RLE-Bench: 48 everyday robotics engineering tasks spanning closed-loop control, policy learning, perception, and mechanical design.

[Lihe Li @lilh76]

🎉 3rd paper as first author! CODI was accepted to #AAMAS2026! We use LLMs and compositional diffusion models to solve offline MARL given datasets with imbalanced agent quality. Hope to see you in Paphos!
Paper: https://t.co/wZiBNNxkWF

[Lihe Li @lilh76]

@_Yelow 😭😭😭

[Lihe Li @lilh76]

Just arrived in beautiful Vancouver! Excited to see you at the West Exhibition Hall B2-B3 on Thursday afternoon! #ICML

[Lihe Li @lilh76]

🎉 Second paper as first author! SemDiv was accepted to #ICML2025 ! We use LLMs to propose novel coordination behaviors and train aligned teammate policies, achieving better diversity & coordination performance! Hope to see you in Vancouver!
Paper: https://t.co/VIBMVExUol

[Lihe Li @lilh76]

@zhaoxuan_t Congrats!!

[Lihe Li @lilh76]

🎉 Second paper as first author! SemDiv was accepted to #ICML2025 ! We use LLMs to propose novel coordination behaviors and train aligned teammate policies, achieving better diversity & coordination performance! Hope to see you in Vancouver!
Paper: https://t.co/VIBMVExUol

[Lihe Li @lilh76]

@JinyiYe6 caught a handsome NJU guy that I know🤣

[Lihe Li @lilh76]

Mission accomplished at #IJCAI2024.
Meeting a lot of great people. What an amazing conference!
(but got hit by covid on the last day😭)

[Lihe Li @lilh76]

@DylanRobertCope @IJCAIconf spotted u at the multi-agent learning workshop😆

[Lihe Li @lilh76]

안녕하세요~
I will attend #IJCAI2024 in Jeju from 8.3 to 8.9🤗
Come check out our poster and pre on 8.7, from 15:00~16:30, at 2F-Room 202A! And feel free to chat with me anytime!
Paper: https://t.co/5QX4Y3Fxg0

[Lihe Li @lilh76]

👏🏻CORe3 is accepted to IJCAI2024!
🤖What happens when the objectives change over time in multi-objective RL? We formalize such situations and proposed CORe3 to solve it!
🇰🇷Looking forward to visit the beautiful Jeju!

[Lihe Li @lilh76]

@AparnaMoumi Hope they are safe.🙏

[Lihe Li @lilh76]

@zhaoxuan_t @ECMLPKDD that's not my name🤣but thanks

[Lihe Li @lilh76]

Our work, Dynamics Adaptive Safe Reinforcement Learning with a Misspecified Simulator, is accepted to @ECMLPKDD'24!
paper: https://t.co/bcfigS8fJr

[Lihe Li @lilh76]

🤣just had a dream: A large AI company redirected its investments from AI-safety to interstellar traveling and used a wild AI to create a world with entirely new physical laws in a distant part of space...The headquarters on Earth was sealed off...Journalists everywhere...

[Lihe Li @lilh76]

@cheryyun_l Honored to have contributed🤣

[Lihe Li @lilh76]

@HaileyJoren Didn't know that ICLR has a Baby Track🤣

[Lihe Li @lilh76]

🎉Exciting times at #AAMAS24! Learned so much and made a lot of fantasic new friends!
👏🏻Ruiqi did an outstanding job presenting our paper COSTA. So proud of him.
🤗Can't wait for the next conference (#IJCAI24 maybe?), and hope to visit beautiful New Zealand again!

[Lihe Li @lilh76]

@1erinrichardson @AAMASconf Learned a lot this morning!