Annabella Macaluso

PhD Student CMU. Prev: BS/MS @UCSDJacobs, @autodesk

[Hyun Joe @hyunjoej]

Excited to share the first paper of my PhD!

If you’ve ever tried to control a VLA via natural language, you know it rarely does what it is told. 🗣️ We introduce a multi-stage pipeline for training a Language Feedback Policy (LFP) to steer a VLA in-the-loop.

[Annabella Macaluso @anna_macalus]

So so cool!

[yisha @yswhynot]

If the end goal of robot hands is to perform human motion, then we should optimize the hardware design with human motion - and on a large scale!
We can generate both a high-dof generalist hand, and also low-dof specialized hands from human demonstration.

[Annabella Macaluso @anna_macalus]

@lltjuatja congrats! 🎉

[Annabella Macaluso @anna_macalus]

@shahdhruv_ Cool demo! Is this behavior emergent from other unrelated randomizations? Or still from randomized table heights (either in sim or real)?

[Adam Hung @Adamjhung]

Learning from human videos often requires restrictive, carefully choreographed human motions.

We propose ✨3PoinTr✨: a scalable way to pretrain from casual human videos. It bridges the embodiment gap by learning 3D scene evolution, enabling learning from natural human motions.

[Annabella Macaluso @anna_macalus]

@DanHrmti @worldmodel_26 I appreciate the tests on zero-shot compositional generalization, looking forward to the code release.

[Annabella Macaluso @anna_macalus]

Congratulations @DanielXieee ! It's super exciting to see your progress and commitment in creating compact, realistic dexterous hand hardware. A lot of skill and good engineering at play here, all the best!

[Quanting Xie @DanielXieee]

A few days ago we got in YC W26, and here is we are working on.

Building hardware is hard, but I really like a quote from @yukez: “People who are really serious about robot learning should make their own robot hardware.”

[Annabella Macaluso @anna_macalus]

@DJiafei @metaai Wow looks cool! Super clean

[Xueyan Zou @xyz2maureen]

As an AI researcher, are you interested in tracking trends from CV/NLP/ML to robotics—even Nature/Science. Our paper “Real Deep Research for AI, Robotics & Beyond” automates survey generation and trend/topic discovery across fields
🔥Explore RDR at https://t.co/uBJ0bYNy3R

[Annabella Macaluso @anna_macalus]

@Stone_Tao @svlevine To your point on starting in sim being unnatural, I think race car drivers who first learned exclusively in sim before transitioning to real races as a fascinating example of sim2real. Although I suppose they already possessed strong physical priors from day to day life.

[Binghao Huang @binghao_huang]

Tactile interaction in the wild can unlock fine-grained manipulation! 🌿🤖✋

We built a portable handheld tactile gripper that enables large-scale visuo-tactile data collection in real-world settings.

By pretraining on this data, we bridge vision and touch—allowing robots to:
✅ Perform robust in-hand reorientation
✅ Control contact and force with precision

🔗 Project page: https://t.co/zcJBEzd23i

(1/6)

[Towaki Takikawa / 瀧川永遠希 @yongyuanxi]

You Can't Manufacture a NeRF - an ICML position paper from CMU that discusses what it means to generate a 3D object that can be manufactured.

Nothing too controversial and feels a bit more like a survey- but good reference to have for explaining this concept 🙂

[Annabella Macaluso @anna_macalus]

@RchalYang @xiaolonw Congrats! Best of luck in your future endeavors

[Annabella Macaluso @anna_macalus]

@Jerry_XU_Jiarui @xiaolonw Congratulations Jiarui! 🎉

[Annabella Macaluso @anna_macalus]

@tennyyin Cool work! Looking forward to the code release.

[Annabella Macaluso @anna_macalus]

@xiaolonw Congratulations! Well deserved

[Xueyan Zou @xyz2maureen]

[1/n] We are releasing M3 (#ICLR2025): a Gaussian Splatting method that builds LMM memories for arbitrary scenes.
🔥 [Efficient] 16 degrees in each Gaussian primitive for one LMM.
🔥 [Alignment] The rendered features are directly in the source LMM embedding space.

[Annabella Macaluso @anna_macalus]

He's just a chill guy.

[Isabella Liu @Isabella__Liu]

🐅 Want to rig your favorite meme character?
Try “RigAnything: Template-Free Autoregressive Rigging for Diverse 3D Assets”!

✨RigAnything is a transformer-based model that sequentially generates skeletons without predefined templates. It creates high-quality skeletons for shapes in any pose, completing rigging in under 2 seconds per shape. 🧵(1/n)

[Annabella Macaluso @anna_macalus]

Great insights Stone, thanks for getting this out so fast

[Stone Tao @Stone_Tao]

Yesterday the hyped Genesis simulator released. But it's up to 10x slower than existing GPU sims, not 10-80x faster or 430,000x faster than realtime since they benchmark mostly static environments

blog post with corrected open source benchmarks & details: https://t.co/7f183ZXVGv

[Annabella Macaluso @anna_macalus]

@Stone_Tao Looking forward to the report 👍