Adam Wei
@adamwei_
Robotics PhD student @MIT_CSAIL
Learning from both sim+real data could scale robot imitation learning. But what are the scaling laws & principles of sim+real cotraining? We study this in the first focused analysis of sim+real cotraining spanning 250+ policies & 40k+ evals arxiv.org/abs/2503.22634 (1/6)
TRI's latest Large Behavior Model (LBM) paper landed on arxiv last night! Check out our project website: toyotaresearchinstitute.github.io/lbm1/ One of our main goals for this paper was to put out a very careful and thorough study on the topic to help people understand the state of the…
Announcing Ambient Diffusion Omni — a framework that uses synthetic, low-quality, and out-of-distribution data to improve diffusion models. State-of-the-art ImageNet performance. A strong text-to-image results in just 2 days on 8 GPUs. Filtering ❌ Clever data use ✅
What if an LLM could update its own weights? Meet SEAL🦭: a framework where LLMs generate their own training data (self-edits) to update their weights in response to new inputs. Self-editing is learned via RL, using the updated model’s downstream performance as reward.
Really cool work on environment generation + inference time search with diffusion!
Want to scale robot data with simulation, but don’t know how to get large numbers of realistic, diverse, and task-relevant scenes? Our solution: ➊ Pretrain on broad procedural scene data ➋ Steer generation toward downstream objectives 🌐 steerable-scene-generation.github.io 🧵1/8
We're really honored to be named alongside Ajay and Justin for this award, from the hands-down best technical committee in robotics! Here's the paper arxiv.org/abs/2304.11259 and accompanying video youtube.com/watch?v=L57Jz3…
[Announcement] This year, we have two (tied) best paper award winners: - Consensus Complementarity Control for Multi-Contact MPC, by A. Aydinoglu, A. Wei, W. Huang, M. Posa - Constrained Articulated Body Dynamics Algorithms, by A. Sathya, J. Carpentier (1/n)
Amazing concurrent work showing the effectiveness of sim+real cotraining on a wide range of tasks! The similarity of our findings suggests that sim data could unlock massive improvements in robotics... I'm excited to see what comes next in this space
How to use simulation data for real-world robot manipulation? We present sim-and-real co-training, a simple recipe for manipulation. We demonstrate that sim data can significantly enhance real-world performance, even with notable differences between the sim and the real. (1/n)