Will Chen

@verityw_

PhD Student @ UC Berkeley RAIL Lab (adv. @svlevine) - MIT SB '22, MEng '23 - 🤖 - 🏹 - 🐕

Joined November 2021

165

Will Chen
@verityw_
Jul 16
Honored to be giving an invited talk at the RL4VLAs workshop at #RSS2026! I'll be discussing how semantic steering enables learning & improving at new tasks for VLAs. 5-5:30p at CB11.00.405, then panel from 5:30-6p. Pull up with some spicy questions!
sites.google.com
RL4VLA Workshop
About
Will Chen
@verityw_
Jul 13
Excited to be at #RSS2026! I'll be presenting on Steerable Policies today at Imit. Learning 1, and giving an invited talk about improving generalist policies with semantic reasoning at the RL4VLAs workshop on Friday. Come say hi if you're interested in embodied robot reasoning!
Will Chen
@verityw_
Jul 3
Generalist robot policies learn diverse skills and behaviors, as steered by language. This enables learning from *semantic* experience, rather than actions. By running RL over prompts, robots can solve new tasks by learning what skills to execute and when! semantic-action-rl.github.io
Jagdeep Bhatia
@JagdeepBhatia8
Jul 3
How can generalist policies adapt to new challenges at deployment using skills they already have? We optimize VLA *prompt inputs* with reinforcement learning, enabling efficient real-robot adaptation on complex tasks where existing methods struggle. 🧵 semantic-action-rl.github.io
00:00
Will Chen
@verityw_
Jun 12
Generalist robot policies learn many useful skills. How can we elicit relevant behaviors when faced with new tasks? We introduce Flow Reversal Steering (FRS): a way to refine coarse actions produced by semantic reasoning into similar precise ones! flow-reversal-steering.github.io 1/N
00:00
Will Chen
@verityw_
Feb 16
How can robot policies be trained to best leverage VLMs' CoT reasoning and in-context learning for generalization? The key is Steerable Policies: vision-language-action models that can be flexibly controlled in many ways! steerable-policies.github.io 1/9
00:00