#ByteDanceSeed - Twitter Hashtag

6 months ago

NEW SOTA 👑 Seed-2.0 Pro is now the #1 model (60.6) on the BabyVision Benchmark, overtaking Gemini-3 Pro (49.7). Huge congrats to the #BytedanceSeed team!

UniPat_AI's tweet photo. NEW SOTA 👑
Seed-2.0 Pro is now the #1 model (60.6) on the BabyVision Benchmark, overtaking Gemini-3 Pro (49.7).

Huge congrats to the #BytedanceSeed team! https://t.co/x8GCAfyKlJ

0

21

5

1

4K

T.Yamazaki @ZappyZappy7

8 months ago

シューズの左右の穴に靴紐を通していく双腕ロボット (ロボットが実環境で複数ステップの高精度タスクを安定して実行できるようにする強化学習フレームワーク) https://t.co/gdJAxKEBLZ #humanoidrobot #BytedanceSeed #GRRL #ReinforcementLearning #framework #dexterous #manipulation #VLA

0

26

3

7

3K

wesley hsieh

@chengyenhsieh

8 months ago

⭐Attending #NeurIPS 2025. Happy to chat about diffusion protein language models (DPLM) from #ByteDanceSeed, and multimodal generative models. Feel free to reach out if you also work in a related space. Our team is also recruiting 2026 summer research interns. Feel free to chat if you are curious about generative models for proteins.

chengyenhsieh's tweet photo. ⭐Attending #NeurIPS 2025.

Happy to chat about diffusion protein language models (DPLM) from #ByteDanceSeed, and multimodal generative models. Feel free to reach out if you also work in a related space.

Our team is also recruiting 2026 summer research interns. Feel free to chat if you are curious about generative models for proteins.

1

3

1

2

436

Xiao Ma

@yusufma555

10 months ago

🚀 New paper alert! We introduce ByteWrist — a compact, anthropomorphic robotic wrist that enables dexterous manipulation in confined spaces. Think home service, medical robots, or precision assembly. 👉 https://t.co/WxBCeAMmpM #ByteDanceSeed #EmbodiedAI #Robotics

7

310

52

128

18K

Xiao Ma

@yusufma555

11 months ago

Thrilled to announce that our team at #ByteDanceSeed has FOUR papers accepted at #NeurIPS2025! 🔥 From novel RL policies to next-gen robotic manipulation and learning from human feedback, we've been busy. A huge thank you to all my amazing collaborators!

2

8

1

0

634

Jean-François Ton @jeanfrancois287

about 1 year ago

I will also be presenting our other works on Active Reward Modelling at #ICML2025 #BytedanceSeed ! Tuesday, 15 July 16:30 - 19:00 East Exhibition Hall A-B #E-2502 Unfortunately there is a time Conflict with another paper so I try to move around :)

Jean-François Ton @jeanfrancois287

over 1 year ago

📢New Paper on Reward Modelling📢 Ever wondered how to choose the best comparisons when building a preference dataset for LLMs? Our latest paper revives classic statistical methods to do it optimally! Here’s a 🧵on how it works 👇 https://t.co/8TlYZCh6ra

jeanfrancois287's tweet photo. 📢New Paper on Reward Modelling📢

Ever wondered how to choose the best comparisons when building a preference dataset for LLMs?

Our latest paper revives classic statistical methods to do it optimally!

Here’s a 🧵on how it works 👇
https://t.co/8TlYZCh6ra https://t.co/9dg1QwWxEz

1

37

9

15

9K

0

4

0

358

Jean-François Ton @jeanfrancois287

about 1 year ago

✈️ I am currently in Vancouver #ICML2025 and will be presenting our latest work on reward models this afternoon ! Tue, 15 July at 4:30 p.m. in East Exhibition Hall A-B, poster E-2504! #BytedanceSeed

Jean-François Ton @jeanfrancois287

about 1 year ago

📢New Paper on Process Reward Modelling 📢 Ever wondered about the pathologies of existing PRMs and how they could be remedied? In our latest paper, we investigate this through the lens of Information theory! #icml2025 Here’s a 🧵on how it works 👇 https://t.co/nzDr4Wrl15

jeanfrancois287's tweet photo. 📢New Paper on Process Reward Modelling 📢

Ever wondered about the pathologies of existing PRMs and how they could be remedied? In our latest paper, we investigate this through the lens of Information theory! #icml2025

Here’s a 🧵on how it works 👇
https://t.co/nzDr4Wrl15 https://t.co/XOAg6C0MVu

6

332

76

302

34K

0

13

2

1

816

Daoguang Zan @zandaoguang

about 1 year ago

Proud to have contributed to Seed-Coder — now the strongest open-source 8B code model. 🥇SWE-bench & Multi-SWE-bench 🥇IOI Surprising power from just 8B params — solid performance across real-world coding tasks. #AI #LLM #OpenSource #Code #ByteDanceSeed

zandaoguang's tweet photo. Proud to have contributed to Seed-Coder — now the strongest open-source 8B code model.
🥇SWE-bench & Multi-SWE-bench
🥇IOI
Surprising power from just 8B params — solid performance across real-world coding tasks.

#AI #LLM #OpenSource #Code #ByteDanceSeed https://t.co/h3Fy7XUs3h

Teortaxes▶️ (DeepSeek 推特🐋铁粉 2023 – ∞)

@teortaxesTex

about 1 year ago

ByteDance Seed shares coding LLMs: Seed-Coder-8B. 6T tokens, outperforms Qwen3-8B. Base, Instruct, Reasoner. The core point: Let the Code Model Curate Data for Itself Immediate SoTA data recipe, far expanding on DeepSeek's early papers. Probably #1 AGI team in China already.

teortaxesTex's tweet photo. ByteDance Seed shares coding LLMs: Seed-Coder-8B. 6T tokens, outperforms Qwen3-8B. Base, Instruct, Reasoner. The core point:
Let the Code Model Curate Data for Itself
Immediate SoTA data recipe, far expanding on DeepSeek's early papers.
Probably #1 AGI team in China already.

8

507

52

343

60K

0

11

1

766

Top Tweets for #ByteDanceSeed

Last Seen Hashtags on Sotwe

Trends for you

Most Popular Users