Hongsuk Benjamin Choi
@redstone_hong
Followers
233
Following
212
Media
16
Statuses
70
RT @shreyasgite: Five Golden Nuggets from this talk:. 1. Pretrained finetuned vs single task policy:.- Because of training on different tas….
0
10
0
RT @hanna_mao: We build Cosmos-Predict2 as a world foundation model for Physical AI builders — fully open and adaptable. Post-train it for….
0
70
0
RT @_wenlixiao: ummm… As a robotics PhD student, I’m genuinely worried that the problem I find important now will be solved in the next 2 y….
0
33
0
Amazing @taeinkwon1.
EgoPressure at @CVPR 2025 (Highlight)!. w/ @_yimzhao_ , @taeinkwon1 , @mapo1 , @cholz . 🌍 Project page: 📰 Paper: 📍Come visit us during our poster session in ExHall D Poster 149, 15 Jun, 16:00 CDT — 18:00 CDT!
0
0
1
RT @seohong_park: Q-learning is not yet scalable. I wrote a blog post about my thoughts on scalable RL algorithms.….
0
186
0
RT @chongyiz1: 1/ How should RL agents prepare to solve new tasks? While prior methods often learn a model that predicts the immediate next….
0
33
0
Cudos to VideoMimic team:).
Great work with super cool avatar reconstruction &animation demos from @w_zielonka ,and humanoid robot mimics from videos @redstone_hong at POETs workshop #CVPR
0
2
17
RT @neerjathakkar: Can we systematically generalize AR "word models" into "world models”? Our CVPR 2025 paper introduces a unified, general….
0
9
0
RT @adcock_brett: Singapore's Sharpa unveiled SharpaWave, a lifelike robotic hand. —Features 22 DOF to balance for dexterity and strength.—….
0
8
0
RT @seohong_park: Is RL really scalable like other objectives?. We found that just scaling up data and compute is *not* enough to enable RL….
0
142
0
RT @Luckyballa: So Flow Matching is *just* . xt = mix(x0, x1, t).loss = mse((x1 - x0) - nn(xt, t)). Nice, here it is in a fragment shader :….
0
66
0
RT @seohong_park: We found a way to do RL *only* with BC policies. The idea is simple:. 1. Train a BC policy π(a|s).2. Train a conditional….
0
41
0
real-to-sim-to-real 🫡.
🧑🤖 Introducing Human2Sim2Robot! 💪🦾. Learn robust dexterous manipulation policies from just one human RGB-D video. Our Real→Sim→Real framework crosses the human-robot embodiment gap using RL in simulation. #Robotics #DexterousManipulation #Sim2Real. 🧵1/7
0
0
5
RT @younggyoseo: Excited to present FastTD3: a simple, fast, and capable off-policy RL algorithm for humanoid control -- with an open-sourc….
0
113
0
RT @TheHumanoidHub: In their latest video, Boston Dynamics’s AI team explains how they make the Atlas humanoid perceive and interact with t….
0
168
0