Ilya Zisman (@suessmannn) 's Twitter Profile
Ilya Zisman

@suessmannn

RL i guess

ID: 1454133208207998979

linkhttps://zis.mn calendar_today29-10-2021 17:09:33

23 Tweet

52 Followers

250 Following

Maksim Zhdanov (@max_nygma) 's Twitter Profile Photo

🔥 Φ-Module 🔥 We present Φ-Module - an easy way to compute partial charges and electrostatic energy in a self-supervised way for any neural network interatomic potential. Learn more below!

🔥 Φ-Module 🔥

We present Φ-Module - an easy way to compute partial charges and electrostatic energy in a self-supervised way for any neural network interatomic potential. Learn more below!
Alexander Nikulin @ ICLR (@how_uhh) 's Twitter Profile Photo

🎥 Pre-training VLAs on human videos is tempting — Latent Action Models quickly become an essential part of leading VLAs, like GR00T (Jim Fan) — but can they effectively handle messy real‐world videos? In our #ICML paper we give an answer: not yet, at least without some help!

🎥 Pre-training VLAs on human videos is tempting — Latent Action Models quickly become an essential part of leading VLAs, like GR00T (<a href="/DrJimFan/">Jim Fan</a>) — but can they effectively handle messy real‐world videos? 

In our #ICML paper we give an answer: not yet, at least without some help!
Denis Tarasov (@ml_is_overhyped) 's Twitter Profile Photo

LLMs are amazing because they can learn in context — read, adapt, and act. Can we do the same for reinforcement learning? That’s the promise of In-Context RL (ICRL). But existing offline ICRL methods don’t even optimize rewards. Our new paper shows why RL matters 🧵

LLMs are amazing because they can learn in context — read, adapt, and act.

Can we do the same for reinforcement learning? That’s the promise of In-Context RL (ICRL).

But existing offline ICRL methods don’t even optimize rewards.

Our new paper shows why RL matters
🧵
Vladislav Kurenkov (@vladkurenkov) 's Twitter Profile Photo

🚀 Introducing cadrille: a new SOTA model for CAD reconstruction from images, point clouds, and text—all in one framework with the use of RLVR. Multimodal inputs + RLVR = clean, editable 3D models. 🧵👇

🚀 Introducing cadrille: a new SOTA model for CAD reconstruction from images, point clouds, and text—all in one framework with the use of RLVR.

Multimodal inputs + RLVR = clean, editable 3D models.

🧵👇
Jay Yang (@jayyanginspires) 's Twitter Profile Photo

As boring as it sounds, I’m slowly realizing that 90% of success is doing the obvious thing for a painfully long amount of time without convincing yourself you’re smarter than you are.

Aaron Wetzler (@aaronwetzler) 's Twitter Profile Photo

Enabling useful teleoperation for complex dexterous robotic arm tasks in decent working spaces proved challenging with the open source SO100 design ($100 arms). Reach, degrees of freedom, and strength were insufficient. The next price point ($1000) felt unnecessary. We’ve been

Alicja Ziarko (@ziarkoalicja) 's Twitter Profile Photo

Can complex reasoning emerge directly from learned representations? In our new work, we study representations that capture both perceptual and temporal structure, enabling agents to reason without explicit planning. princeton-rl.github.io/CRTR/