Yuda Song

@yus167

PhD @mldcmu. Previously @ucsd_cse @UcsdMathDept

yudasong.github.io

Joined April 2020

Replies Reposts Media

Pinned
Yuda Song
@yus167
Oct 15, 2025
🤖 Robots rarely see the true world's state—they operate on partial, noisy visual observations. How should we design algorithms under this partial observability? Should we decide (end-to-end RL) or distill (from a privileged expert)? We study this trade-off in locomotion. 🧵(1/n)
2
Yuda Song
@yus167
Sep 2
The most common question we get after MaxRL is: how do you handle continuous reward? Now I think we have a good answer to that -- check out the post from @stablegradients to see how we did it 👇 (We'll share more evidence for why we believe this is the correct extension soon.)
Shrinivas Ramasubramanian
@stablegradients
Sep 2
Is RL optimizing the right objective? 🤔 Should we maximize mean reward? Best-of-k? Which k? Standard RL pulls on the mean and often the distribution collapses to a spike. The tail dies 🥲 We introduce Tail-Likelihood Reinforcement Learning (TailRL). It maximizes the mean
00:00
1
Yuda Song
@yus167
Jul 3
I will be at ICML 🇰🇷 from Mon to Sat, and present two posters and one oral (more in the thread). Look forward to connecting with friends and discussing RL research (LLMs, robotics, theory), self-improvement, and continual learning.
8
Yuda Song
@yus167
May 19
Exciting work! But in our February paper, "Reinforcement Learning with Text Feedback", we proposed the same methodology: predicting environment feedback on top of the RL loss. Nice to see this idea specialized to agentic terminal tasks, and the new insight this brings 💡. [1/2]
Dimitris Papailiopoulos
@DimitrisPapail
May 18
Article
ECHO: Terminal Agents Learn World Models for Free
Co-written with @VaishShrivas We taught CLI agents to predict terminal responses during RL, alongside the usual GRPO loss on actions. The change is tiny: same rollout and forward pass, but stop...
3
Yuda Song
@yus167
May 9
Congrats to my only labmate (in Drew's lab)!
Gokul Swamy
@g_k_swamy
Apr 20
it took a minute, but i'm proud to share that i'm finally "Dr. Swamy" :)
1