Yaoyao(Freax) Qian
@RubyFreax
🔬 RA @HelpingHandsLab @ischoolui | Ex MLE Coop@ Lendbuzz | Ex Technical Manager | Full-Stack Developer | 🤖 | 🦾 | LLM | 🐱🐶🐯🦁🦋
🥰Now is a #CoRL2024 Paper!
❓ How do you solve grasping problems when your target object is completely out of sight? 🚀 Excited to share our latest research! Check out ThinkGrasp: A Vision-Language System for Strategic Part Grasping in Clutter. 🔗 Site: h-freax.github.io/thinkgrasp_page
Finally cleaned up my messy code—now easy to reproduce! 🎉 Current version uses langsam (no dual 3090s for vlpart setup yet [😭]). Modules are flexible for swapping or optimizing (e.g., better models for segmentation 👀). url: github.com/H-Freax/ThinkG…
![RubyFreax's tweet image. Finally cleaned up my messy code—now easy to reproduce! 🎉
Current version uses langsam (no dual 3090s for vlpart setup yet [😭]). Modules are flexible for swapping or optimizing (e.g., better models for segmentation 👀).
url: github.com/H-Freax/ThinkG…](https://pbs.twimg.com/media/Ge4vf1cWkAEVCoh.jpg)
Excited to share our #ICML2025 paper, Hierarchical Equivariant Policy via Frame Transfer. Our Frame Transfer interface imposes high-level decision as a coordinate frame change in the low-level, boosting sim performance by 20%+ and enabling complex manipulation with 30 demos.
Owen will be presenting our poster for the paper Hierarchical Equivariant Policy via Frame Transfer at ICML Today (see lnkd.in/e-7p9Viq for details). If you are interested in equivariance and/or robotic manipulation please stop by!
🥳Visual Tree Search of Web Agent has been accepted!
🎉 Exciting News! We're thrilled to announce that our paper "Visual Tree Search of Web Agent" has been accepted to ECML-PKDD 2025, one of the premier European conferences in machine learning and data science! This breakthrough work comes from our talented PathOnAI.org…
🔥 We introduce Multiverse, a new generative modeling framework for adaptive and lossless parallel generation. 🚀 Multiverse is the first open-source non-AR model to achieve AIME24 and AIME25 scores of 54% and 46% 🌐 Website: multiverse4fm.github.io 🧵 1/n
📢 (1/16) Introducing PaTH 🛣️ — a RoPE-free contextualized position encoding scheme, built for stronger state tracking, better extrapolation, and hardware-efficient training. PaTH outperforms RoPE across short and long language modeling benchmarks arxiv.org/abs/2505.16381
🎉 We’re excited to host two challenges at LOVE: Multimodal Video Agent Workshop at CVPR 2025, advancing the frontier of video-language understanding! @CVPR #CVPR2025 📌 Track 1A: [VDC] Video Detailed Captioning Challenge Generate rich and structured captions that cover multiple…
🤖 How do AI agents actually work together? I made 2 short videos on Google’s Agent2Agent (A2A) protocol: 📘 Ep1: What is A2A? 📙 Ep2: Why it matters No backend needed—just curiosity. 🎥 Watch here: youtube.com/playlist?list=…
Just posted a 21-min tutorial on Model Context Protocol (MCP) — no jargon, just real-life analogies. 🍜 Restaurant menus 🧳 Travel guides 🦸♂️ Superpowers 📝 Memory notes I wanted to make it clear enough for anyone, even without a tech background. 🎥👇 youtu.be/0EtVAzIYbys?si…
Just realized my paper is being used as a baseline—such a strange feeling! Seeing my model tested across different settings without me doing anything is fascinating. 🤯 Added the papers using ThinkGrasp as a baseline to its GitHub—check them out!🥳
Introducing TraceVLA: a fully open-source Vision-Language-Action model reimagining spatial-temporal awareness: tracevla.github.io ✨ 3.5x gains on real robots, SOTA in simulation 💡 Fine-tunes on just 150K trajectories ⚡ Compact 4B model = 7B performance
Mitigating racial bias from LLMs is a lot easier than removing it from humans! Can’t believe this happened at the best AI conference @NeurIPSConf We have ethical reviews for authors, but missed it for invited speakers? 😡