Fox_block (@Fox__Block) header
Fox_block (@Fox__Block) profile picture

minecraft builder and render artist ||open for render commissions discard: foxblock#7406 artstation: https://nitter.gq/t.co/VivJJPwjxP

Joined February 2016
Photo by Fox_block (@Fox__Block)Photo by Fox_block (@Fox__Block)Photo by Fox_block (@Fox__Block)Photo by Fox_block (@Fox__Block)Photo by Fox_block (@Fox__Block)Photo by Fox_block (@Fox__Block)Photo by Fox_block (@Fox__Block)Photo by Fox_block (@Fox__Block)Photo by Fox_block (@Fox__Block)Photo by Fox_block (@Fox__Block)
Fox_block retweeted
🎉 Robots in the real world have to move around. Teleoperating a robot to do both at once is still slow and awkward Today we're excited to release HALO, Human-Aligned Data Collection for Dexterous Whole-body Mobile Manipulation! HALO lets you hold the robot's grippers directly and just walk, and the platform rolls with you 🚶 You get natural whole-body motion and fine bimanual control at the same time. Three key results: 🔑 Demo collection approaching natural human speed: operators walk at a normal pace (~0.7 m/s), up to 2.9× faster than Mobile ALOHA with comparable or higher success 🔑 Dexterous skills: the policy model trained on HALO data fastens a short zip tie and disassembles Lego, while policies trained on Mobile ALOHA or VR data fail 🔑 Mobile skills: very high success on a six-stage long-horizon mobile manipulation task Let's check out our demos 👇
9
7
2
82
4,348
Fox_block retweeted
𝗠𝗮𝗸𝗶𝗻𝗴 𝗱𝗲𝗺𝗼𝘀 𝗹𝗼𝗼𝗸 𝗹𝗶𝗸𝗲 𝗱𝗲𝗽𝗹𝗼𝘆𝗺𝗲𝗻𝘁 🪄 Walking with the robot is natural, but the data still has to match what the robot sees and does at test time. 🔹 Base motion: two encoders track the platform, and we map their readings to the base's linear and angular velocity, then shift them in time to account for the base's response delay 🔹 Vision: we detect the operator's hands (CaRE-Ego), track them with SAM2 and inpaint them away with DiffuEraser Removing the hands alone lifts Push Chair success from 55% to 85%.
1
1
1
2
331
Fox_block retweeted
𝗗𝗮𝘁𝗮 𝗰𝗼𝗹𝗹𝗲𝗰𝘁𝗶𝗼𝗻 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆 📈 We compared HALO with Mobile ALOHA and VR teleoperation on Push Chair and Disassemble Lego: same operator, same tasks, same time budget, and the operator had never used any of the three before. 🔹 vs Mobile ALOHA: 2.8–2.9× faster per successful demo, with equal or higher success 🔹 vs VR: 1.6–2.7× faster, with 4–5× higher success VR makes precise gripper control hard, while Mobile ALOHA's powered base makes the operator pause before grasping the chair.
1
1
2
223
Fox_block retweeted
𝗪𝗵𝗮𝘁 𝘁𝗵𝗲 𝗽𝗼𝗹𝗶𝗰𝗶𝗲𝘀 𝗹𝗲𝗮𝗿𝗻 🤖 We fine-tune π0.5 per task on HALO data. It takes head and wrist cameras and outputs arm joints plus base velocity, using asynchronous real-time action chunking. 🔹 Push Chair 85%: handles 2–3 m changes in start pose and rooms it has never seen 🔹 Disassemble Lego 75% (30/40), adapting its strategy to randomized structures 🔹 Zip-Tie Fastening 30%, where Mobile ALOHA and VR data give 0% 🔹 Long-horizon (shelf → load → lift → drive → conveyor → return): 73.3% (11/15) With the same collection budget, policies trained on Mobile ALOHA or VR data fail on Lego and the long-horizon task.
1
1
1
247
Fox_block retweeted
𝗣𝗮𝗽𝗲𝗿 & 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗽𝗮𝗴𝗲 🔗 📄 Paper: arxiv.org/abs/2611.16439 🌐 Project page: robbyant-research.github.io/… 💻 Code: github.com/robbyant-research… Huge thanks to the team from HKUST and Robbyant, led by Shuai Yang and Fangyi Xu!
1
3
229
I render a new map from @justadoubi in blender. ps: I am open to accepting rendering commissions. #Minecraft #minecraft建築コミュ #Minecraftbuilds
21
729
🚀 Exciting news! @intern_lm 7/20B models are now live on the @huggingface Open LLM Leaderboard! 🔍 Highlights: - 200K context length for base/chat models. - 20B model is on par with the performance of Yi-34B. - 7B model is the best in the <= 13B range. huggingface.co/internlm
3
32
5
137
35,015
Fox_block retweeted
2
16
100
6,747
我想我终于完成这台[王狩]了,它还有更多东西需要展示 #minecraft #Minecraftbuilds #minecraft建築コミュ
3
18
116
7,583