南田桂吾 Keigo MINAMIDA

UTokyo @ishi96 lab D1（KINDAI Mechatronics lab → UTokyo @rkmt lab）｜Sony CSL リサーチアシスタント｜HCI・Humanoid｜セルフツアー・安全走行アプリwakabarを開発しています

[Google DeepMind @GoogleDeepMind]

Gemini Robotics 2.5 is here. 🤖

Our latest physical AI is able to learn hundreds of everyday skills - from folding laundry and serving pastries to autonomously sorting car parts on the factory floor.

Here’s a snapshot of its capabilities 🧵

[南田桂吾 Keigo MINAMIDA @keigominamida]

パーソナルAIエージェント「Muse」をチェックしてみよう。参加後48時間以内に[設定]で私のコードを利用すると、私たち両方が10億のMuseトークンを獲得できます。

コード: 9QBEQF
https://t.co/FSqwFrUw7V

[NVIDIA Robotics @NVIDIARobotics]

A finer touch for humanoid robots. 🤖

At #CoRL2026, NVIDIA is introducing GR00T 2, an open world action model for humanoids, alongside new open industrial benchmarks from @RLWRLD_ai and @UCBerkeley built on Isaac Lab-Arena.

Explore how NVIDIA’s open world action models, scalable simulation tools and agentic workflows help researchers train, test and refine robot skills, from connecting cables to assembling small components.

Discover more ➡️ https://t.co/spzxPurj6y

[南田桂吾 Keigo MINAMIDA @keigominamida]

もう2か。早い。

[Yuke Zhu @yukez]

Meet GR00T 2, our new generation of world-action models (WAMs): it imagines how the physical world will unfold and generates the actions that get it there. One model runs on parallel grippers, dexterous hands, and humanoids.

https://t.co/32Z1plSGQg

[Claude @claudeai]

Introducing Claude Fable 5.2 and Claude Mythos 5.2, designed to tackle the most daunting technical challenges.

They’re capable collaborators on ambitious projects in many fields, and show improvements in areas like coding, reasoning, and vision.

[南田桂吾 Keigo MINAMIDA @keigominamida]

@NVIDIAGeForceJP めっちゃかっこいい🪄

#RTXON

[NVIDIA GeForce JP @NVIDIAGeForceJP]

DLSS 5 の対応を記念して、『ホグワーツ・レガシー』をテーマにしたカスタム GeForce RTX 5090 をプレゼント🎁

応募方法:
🟢 このアカウントをフォロー
🟢 投稿をシェア
🟢 #RTXON とコメント

応募規約: https://t.co/Mlc84zSurL

[Robotics Papers @OWW]

Warping the Workspace: Expanding Visual Access with Adjustable Reach Mapping for Humanoid Teleoperation

Keigo Minamida, Koya Dendo, Yuto Nakamura, Jun Rekimoto
https://t.co/gQUUA51iHn [𝚌𝚜.𝚁𝙾 𝚌𝚜.𝙷𝙲]

[koya dendo @colloidoooooooo]

#UIST2026 Poster | Tue, Nov 3 | 6:00–9:00 PM 👻
Ghost-in-the-Loop: Can People Distinguish Human and AI Agency in Humanoid Teleoperation?

Can you tell whether a humanoid robot is controlled by a human or an AI?
We switch between human teleoperation and AI-generated voice, facial expressions, and gestures within the same humanoid robot.

Come see our poster and share your thoughts!

[AGIBOT @AGIBOTofficial]

Robots Never Clock Out | AGIBOT G2’s 24-Hour Convenience Store Challenge

Six AGIBOT G2 take on a 24-hour shift as convenience store trainees.
From rush-hour crowds to late-night solo duty and the morning breakfast sprint, can they keep up with nonstop customers and unexpected requests?

https://t.co/eg1fUhcr3T

[南田桂吾 Keigo MINAMIDA @keigominamida]

#UIST2026 posters Tue, 3 Nov | 6:00 PM - 9:00 PM
Warping the Workspace: Expanding Visual Access with Adjustable Reach Mapping for Humanoid Teleoperation

We warp its 360° view and remap arm movements together, so operators can reach to the sides and back while facing forward.

I'd love to hear your thoughts — come chat with us!

[南田桂吾 Keigo MINAMIDA @keigominamida]

テレコスプレ

[koya dendo @colloidoooooooo]

Presenting our poster at #UIST2026! 👻

Can you tell whether a humanoid robot is controlled by a human or an AI?

Ghost-in-the-Loop lets a human operator and an AI switch control of the same humanoid robot.

By recreating the operator's voice, facial expressions, and gestures, we explore whether people can tell the difference—and what makes us perceive someone as human.

-------

UIST 2026で、ヒューマノイドとAIに関する研究をポスター発表します！👻

目の前のヒューマノイドを操作しているのが人間なのかAIなのか、あなたは見分けられるでしょうか？

Ghost-in-the-Loopは、人間による遠隔操作と、本人の声・表情・ジェスチャを再現するAIによる操作を、同じヒューマノイド上で切り替えるシステムです。

[南田桂吾 Keigo MINAMIDA @keigominamida]

Our collaborative work will also be presented at #UIST2026! 👻

Can people tell whether a humanoid robot is controlled by a human or AI?

ーー

こちらの研究も UIST2026 で発表します！

ヒューマノイドを動かしているのは人間か、AIか。人はその違いを見分けられるのでしょうか？

[Jun Rekimoto: 曆本純一 @rkmt]

Can People Distinguish Human and AI Agency in Humanoid Teleoperation? A Preliminary Study of Agency Perception, Xian Li, Koya Dendo, Keigo Minamida, Yuto Nakamura, Per Ola Kristensson, Jun Rekimoto
https://t.co/qJxfnQxYdr
https://t.co/as9GyYZd9z
人は、ヒューマノイドの遠隔操作において、動作の主体が人間かAIかを識別できるのだろうか。この問いを検討するため、ロボットの身体性を一貫して保ちながら、音声・表情・ジェスチャを人間による操作とAIによる生成の双方で制御できる遠隔操作フレームワーク Ghost-in-the-Loop を開発した。

[南田桂吾 Keigo MINAMIDA @keigominamida]

Presenting our poster at #UIST2026! 🤖

Can we reach behind a humanoid robot without turning around?

We warp its 360° view and remap arm movements together, so side and rear targets can be reached from the operator’s forward workspace.

ーー

UIST 2026で、ヒューマノイドの遠隔操作に関する研究をポスター発表します！

360°映像の見せ方と腕の動きを一緒に変換することで、ロボットの側方・後方の空間を正面から操作できるようにする研究です。

[Jun Rekimoto: 曆本純一 @rkmt]

Warping the Workspace: Expanding Visual Access with Adjustable Reach Mapping for Humanoid Teleoperation, Keigo Minamida, Koya Dendo, Yuto Nakamura, Jun Rekimoto, ACM UIST 2026 
https://t.co/RPv44Hzeeb
従来の一人称視点によるヒューマノイドの遠隔操作では、ロボットの視界と操作系を一対一で人間の操作空間にマップするため、側方や後方の対象へのアクセスに操作者自身の回転を要した。本研究では、パノラマ映像の方位角変換と、その逆変換に基づく上肢のヨー角変換を組み合わせ、現実とのマッピングを変容・拡張する視覚–運動連動リマッピング(Reality Remapping)を提案する。

[南田桂吾 Keigo MINAMIDA @keigominamida]

あ、Claudeは10/8からプロモーションクレジットとして毎月200ドル分のAPIくれることやってたのか

[南田桂吾 Keigo MINAMIDA @keigominamida]

Rokid Color 気になるな
https://t.co/7Nz0i8lMDk

[南田桂吾 Keigo MINAMIDA @keigominamida]

なぜこんなに気前いいんだ

[OpenAI Developers @OpenAIDevs]

Starting today, eligible ChatGPT Pro and Business subscribers can claim up to $500 per month in API credits to build apps and agents.

[南田桂吾 Keigo MINAMIDA @keigominamida]

めっちゃ良くなった

[Even Realities @EvenRealities]

Even G2 just got a major AI upgrade.

Version 2.4.0 makes your Even G2 more intelligent, personalized, and useful:
• Knowledge Base (Beta)
• Even AI 2.0
• Conversate Pro Mode (Beta)

Here's the rundown: 🧵

[Google DeepMind @GoogleDeepMind]

This is Genie 4, our most realistic and interactive world model yet. 🌐

Explore dynamic environments in real time to see how the scene responds, sounds, and how the people in it move, react, and speak.