Shubham Nagar

@MarioChan2002 Is it just for VLAs? Did you guys also try world modes in this benchmark?

We evaluated 30+ frontier embodied AI models. The result is clear: current generalist robot policies are still far from robust real-world manipulation. This is why we built RoboDojo.

@LeRobotHF This is great news, just as I was starting to experiment with JEPA based world models. LeRobot released it on their platform.

LeRobot v0.6.0 is officially here: Imagine, Evaluate, Improve! 🤖🚀 We are closing the robot learning loop with massive upgrades for the open-source robotics community. From policies that imagine the future to a much leaner installation, here is what is new: - 🌍 World Models: VLA-JEPA, LingBot-VA, and FastWAM help your policies anticipate the future. - 👀 VLA Expansion: Welcome GR00T 1.7, MolmoAct2, EO-1, Multitask DiT, and EVO1. - 🏅 Reward Models API: Track success seamlessly with Robometer and TOPReward. - 🎯 Unified Evaluation: 6 new simulation benchmarks, all accessible via the lerobot-eval CLI. - ☁️ And more: lerobot-rollout CLI for DAgger corrections, HF Jobs cloud training, up to 2x faster data loading, GUI - LeLab, many docs improvements Ready to build the future of robotics? Dive into the full release notes here: huggingface.co/blog/lerobot-r… @ClementDelangue @Thom_Wolf

Excited to start my work of using KUKA iiwa robot for disassembly automation.

@mweinbach Can it make sense on an enterprise level to do so? You can run other models too on enterprise level GPUs and support a team of developers.

The minimum to run the model is ~$20K in hardware and you get ~20 tok/s out ~$20K gets you around 34.6B tokens at a 12:1 input to output ratio assuming good token caching If you ran the hardware 24/7, it would take roughly 5.5 years to break even

@bha_51 Very insightful!

In order to make a better predictor for a cell's state to perturbation, you need to give the model information on how changing a specific gene X leads to downstream changes Y. This is exactly what CRISPR knockdown experiments capture at the genome scale and datasets like Replogle et al., CD4 Perturb-seq etc exist... What i was trying out today was in addition to this precise but scarce dataset are there more abundant datasets with even faint signals. The place i went looking for this is an eQTL dataset. This dataset contains expression vectors of genes in a cell at population scale. In a population there are random changes in an inherited DNA of individuals that cause changes in the expression of genes. So using this dataset u can effectively synthesize a pseudo-perturbation dataset that measures the downstream effects on genes Y, caused by expression changes in gene X which in turn is caused by an a variation in the DNA from different people. So for testing whether this data has any meaningful signal, I drew the statistics off of CD4 T cell, this was chosen coz there exists both the CRISPR knockdown experiments and also large scale population genetics. After the actual measurement : Natural ripple: split the donors by what type of DNA variant they inherited (that also controls the expression of gene X), and for every gene Y measure how much Y's expression shifts between different groups Engineered ripple: CRISPR-knock-down directly X, measure the change of every gene Y. Ideally what I hoped was that the natural ripple had some correlation with the engineered ripple, but the data shows pretty much no direct correlation above the noise floor. I further tried is there any perceivable signal in the eQTL perturbation that is consistant with itself. I did this by splitting the donors in half and trying to see if there is correlation with the eQTL dataset itself and again nothing over the noise floor. Finally I tried to see is it just because of dilution that I was not seeing correlation. The natural ripple covers ~10,000 genes, and almost all of them have no real connection to X, that's 10,000 noise values drowning the few dozen genes that actually matter. Maybe the signal is real but washed out. So i tried letting CRISPR tell which genes are the real targets, then look at the natural effect only there.... And again no signal only noise floor. So looking at population data to augment targeted data seems to be not working... Need to find something else

@Saanvi_dhillon Hein? I still use it daily lol

Bro disappeared like he never existed.

@Samaytwt Outlook

Name a better alternative to Gmail i’ll wait.

@muskaan___07 Update to Ubuntu 24.04 atleast!

Just installed Linux for the first time What’s the first thing I should install or set up ?

@SanSanychUA Thanks for the reply. I did not test it on real-world industrial scenarios. This was research project done at the institute which is making their own software for robotic simulation and wanted to have something like this implemented in their own software.

@_ok_cc_ Adaptive LiDAR transmission policy for MRG SLAM is exactly what industrial robotics needs. Bandwidth drops in real-world facilities always break multi-robot graphs. Did you test it on dynamic industrial scenes or mostly static environments?

@trikcode I have been wanting to build something as well but I’m not sure where to start or how to deploy it.

honest question: is anyone actually making money from these weekend AI projects or are we all just building for localhost and Twitter screenshots

@ScholarshipfPhd So true 🤣

PhD Supervisor His ideas vs my brain

@bha_51 Congrats!

🚀🚀

@svlevine That’s really inspiring work, I have fun reading about latest research in robotics. My current interest is in using FF-JEPA world models and using RL agents to plan it in the latent space. I have been talking to a professor about this at my university!

Flow reversal steering allows "steering" diffusion-based VLAs with high-level actions, for example from VLM reasoning. This also lets us run RL in the diffusion noise space with exploration guided by high-level reasoning: think through a task, then practice it! 👇

@gowthami_s I would love the online videos of this course too. Do you know if they are creating a MOOC out of it?

I was looking forward to the online lectures of this course. Very timely curriculum

Post a picture YOU took. Just a pic. No description

Train a V-JEPA2 based world model on the push T problem videos. Then used a CEM planner to solve the task. It is able to solve short tasks but not longer tasks. Next step is to train world model to predict more and use RL for planning. #AI #JEPA

@fayezsalka This is very impressive, I would love to know the technical details of how you implemented that.

I wonder what those arms are capable of 🤔 (Realtime and autonomous)

@JunaidAckroyd I am ambitious about embodied robotics.

Reply and I’ll follow you. I want to connect with ambitious people building something great.

@TTrimoreau OpenAI

Name a startup that completely changed an industry

@theo @drisspg I guess he mean 0 should be on the left like normal mathematical charts.

@drisspg ? Top right is good. Bottom left bad. Top left smart but expensive. This is totally normal?

Holy chart crime