Younggyo Seo
@younggyoseo
Followers
880
Following
1K
Statuses
278
Postdoc @ UC Berkeley w/@pabbeel | ex- RS at Dyson w/ @stepjamUK | Ph.D in KAIST
Berkeley, CA
Joined October 2010
Introducing CoordTok, a scalable video tokenizer that can encode a 128-frame video into only 1k tokens. CoordTok learns a mapping from (x, y, t) coordinates to the corresponding patches of input videos. 🧵[1/6] project page:
4
52
219
This is a really well-written, nice paper on using flow matching for RL, you should check this!
Excited to introduce flow Q-learning (FQL)! Flow Q-learning is a *simple* and scalable data-driven RL method that trains an expressive policy with flow matching. Paper: Project page: Thread ↓
1
4
55
RT @seohong_park: Excited to introduce flow Q-learning (FQL)! Flow Q-learning is a *simple* and scalable data-driven RL method that trains…
0
140
0
RT @KyleStachowicz: R1's RL findings are great news for reasoning but grim for robotics. All the major takeaways (ground-truth reward, grea…
0
22
0
RT @chelseabfinn: Disappointed with your ICLR paper being rejected? Ten years ago today, Sergey and I finished training some of the first…
0
169
0
RT @animesh_garg: This is so ridiculous! First figure out a way to built a humongous backlog of (mostly non-white) people in work visa stuc…
0
2
0
RT @yoonholeee: Excited to share our new work on test-time alignment! We introduce HyRe, a fast way to adapt large models (like LLM reward…
0
59
0
RT @rtk254: Video models != world models "We find that across a range of current models (Sora, Runway, Pika, Lumiere, Stable Video Diffus…
0
129
0
RT @ma_nanye: Inference-time scaling for LLMs drastically improves the model's ability in many perspectives, but what about diffusion model…
0
89
0
RT @carlo_sferrazza: Big news for open-source robot learning! We are very excited to announce MuJoCo Playground. The Playground is a repro…
0
20
0
RT @kevin_zakka: The ultimate test of any physics simulator is its ability to deliver real-world results. With MuJoCo Playground, we’ve co…
0
156
0
RT @KarlPertsch: Excited to release FAST, our new robot action tokenizer! 🤖 Some highlights: - Simple autoregressive VLAs match diffusion…
0
95
0
RT @RoboticsSciSys: 📣 "Soft" deadline extension! As you know, RSS 2025 will be held in Los Angeles this June. The entire organizing team, i…
0
19
0
RT @carlo_sferrazza: Ever wondered what robots 🤖 could achieve if they could not just see – but also feel and hear? Introducing FuSe: a re…
0
57
0
RT @oier_mees: Can generalist robot 🤖 policies understand heterogeneous multimodal sensors? Introducing FuSe, a novel finetuning recipe all…
0
23
0
RT @douwekiela: I’m really sad that my dear friend @FelixHill84 is no longer with us. He had many friends and colleagues all over the world…
0
92
0
RT @allenzren: HNY! Lately I took a crack at implementing the pi0 model from @physical_int PaliGemma VLM (2.3B fine-tuned) + 0.3B "action…
0
59
0
RT @kchonyc: feeling a bit under the weather this week … thus an increased level of activity on social media and blog:
0
116
0
RT @huiwon0516: Excited to share CoordTok, a scalable video tokenizer that learns a mapping from coordinates to pixels! With CoordTok, we e…
0
2
0
We believe our work will facilitate future research in designing efficient and scalable video tokenization and generation. This work is led by @huiwon0516 at @kaist_ai , with @sihyun_yu , Jinwoo Shin, and @pabbeel (@berkeley_ai ) 🧵[6/6] paper:
0
0
3