Sateesh Kumar Profile
Sateesh Kumar

@sateeshk21

Followers
398
Following
1K
Media
29
Statuses
107

Computer Vision, Robotics | CS PhD student at @UTAustin | Previous: Researcher @TikTok_US MS CS @UCSanDiego

San Diego, California
Joined August 2019
Don't wanna be here? Send us removal request.
@sateeshk21
Sateesh Kumar
1 month
Which data is best for training few-shot imitation policies for robot manipulation? Some think it’s the data that looks similar, or has similar motion, or comes with related language labels. They are all right AND wrong: depending on the task, sometimes this similarity helps but
1
4
11
@ShivinDass
Shivin Dass
18 days
A little late to this but excited to share that DataMIL won the best paper at the Data workshop at #CoRL! If you haven't already, check it out! 👇
@ShivinDass
Shivin Dass
5 months
Ever wondered which data from large datasets (like OXE) actually helps when training/tuning a policy for specific tasks? We present DataMIL, a framework for measuring how each training sample influences policy performance, hence enabling effective data selection 🧵
1
4
13
@geopavlakos
Georgios Pavlakos
28 days
Sateesh (@sateeshk21) will be presenting COLLAGE at #CoRL2025 today! Check out his presentation and poster (Spotlight 4 & Poster 2)! Project page & code:
@sateeshk21
Sateesh Kumar
1 month
Which data is best for training few-shot imitation policies for robot manipulation? Some think it’s the data that looks similar, or has similar motion, or comes with related language labels. They are all right AND wrong: depending on the task, sometimes this similarity helps but
0
2
8
@sateeshk21
Sateesh Kumar
28 days
I am presenting COLLAGE 🎨 at @corl_conf today. Spotlight presentation: 3:30 pm Poster: 4:30 - 6:00 pm. Poster #41. COLLAGE 🎨 is a data curation approach that automatically combines data subsets selected using different metrics, by weighting each subset based on its relevance
@sateeshk21
Sateesh Kumar
1 month
Which data is best for training few-shot imitation policies for robot manipulation? Some think it’s the data that looks similar, or has similar motion, or comes with related language labels. They are all right AND wrong: depending on the task, sometimes this similarity helps but
0
1
1
@sateeshk21
Sateesh Kumar
1 month
📝 Paper: https://t.co/VjKzPvWoQy 🌐 Project page & code: https://t.co/dzCmNpks4c Work done with my wonderful collaborators at @UTAustin: @ShivinDass, @geopavlakos and @RobobertoMM. Excited to present COLLAGE 🎨 at @corl_conf very soon — Spotlight 4 & Poster 2 on Sept 29!
0
0
2
@sateeshk21
Sateesh Kumar
1 month
We evaluate COLLAGE 🎨 on 10 simulated and 6 real-world tasks. It consistently outperforms single-modality retrieval baselines, showing that adaptively combining subset retrieved using different features leads to more effective retrieval for few-shot imitation learning.
1
0
2
@sateeshk21
Sateesh Kumar
1 month
The predicted weights indicate how informative each modality’s data is for the task. COLLAGE 🎨 uses them to guide mini-batch sampling, with more relevant data chosen more often.
1
0
2
@sateeshk21
Sateesh Kumar
1 month
To predict a subset’s weight, COLLAGE 🎨 trains a lightweight reference policy on each retrieved subset. It then measures task relevance by computing the log-likelihood of target actions under this policy. This strategy is feature-agnostic and avoids needing costly environment
1
0
2
@sateeshk21
Sateesh Kumar
1 month
Given a few target demonstrations, COLLAGE 🎨 works as follows:  1️⃣ Retrieve subsets using different retrieval strategies (eg. visual, motion or language similarity) 2️⃣ Predict a task-specific weight for each retrieved subset 3️⃣ Sample batches using a weighted mixture of
1
0
2
@sateeshk21
Sateesh Kumar
1 month
The predicted weights indicate how informative each modality’s data is for the task. COLLAGE 🎨 uses them to guide mini-batch sampling, with more relevant data chosen more often.
0
0
0
@sateeshk21
Sateesh Kumar
1 month
To predict a data subset’s weight, COLLAGE 🎨 trains a lightweight reference policy on each retrieved subset. It then measures task relevance by computing the log-likelihood of target actions under this policy. This strategy is feature-agnostic and avoids needing costly
1
0
0
@sateeshk21
Sateesh Kumar
1 month
Given a few target demonstrations, COLLAGE 🎨 works as follows: 1️⃣ Retrieve subsets using different retrieval strategies (eg. visual, motion or language similarity) 2️⃣ Predict a task-specific weight for each retrieved subset 3️⃣ Sample batches using a weighted mixture of the
1
0
0
@sateeshk21
Sateesh Kumar
1 month
The predicted weights indicate how informative each modality’s data is for the task. COLLAGE 🎨 uses them to guide mini-batch sampling, with more relevant data chosen more often.
0
0
0
@sateeshk21
Sateesh Kumar
1 month
To predict a data subset’s weight, COLLAGE 🎨 trains a lightweight reference policy on each retrieved subset. It then measures task relevance by computing the log-likelihood of target actions under this policy. This feature-agnostic strategy avoids needing costly environment
1
0
0
@sateeshk21
Sateesh Kumar
1 month
Given a few target demonstrations, COLLAGE 🎨 works as follows: 1️⃣ Retrieve subsets using different retrieval strategies (eg. visual, motion or language similarity) 2️⃣ Predict a task-specific weight for each retrieved subset 3️⃣ Sample batches using a weighted mixture of the
1
0
0
@sateeshk21
Sateesh Kumar
1 month
Excited to share our work MimicDroid. Humanoid learns by watching human videos!
@rutavms
Rutav
1 month
Intelligent humanoids should have the ability to quickly adapt to new tasks by observing humans Why is such adaptability important? 🌍 Real-world diversity is hard to fully capture in advance 🧠 Adaptability is central to natural intelligence We present MimicDroid 👇 🌐
0
0
1
@sateeshk21
Sateesh Kumar
5 months
Shivin and team show how to turn data curation for behavior cloning into a supervised learning problem using DataModels. Clean formulation and strong results!
@ShivinDass
Shivin Dass
5 months
Ever wondered which data from large datasets (like OXE) actually helps when training/tuning a policy for specific tasks? We present DataMIL, a framework for measuring how each training sample influences policy performance, hence enabling effective data selection 🧵
0
0
3
@rutavms
Rutav
1 year
🤖 Want your robot to grab you a drink from the kitchen downstairs? 🚀 Introducing BUMBLE: a framework to solve building-wide mobile manipulation tasks by harnessing the power of Vision-Language Models (VLMs). 👇 (1/5) 🌐 https://t.co/61eev1Jyvw
7
38
174
@sateeshk21
Sateesh Kumar
1 year
Super impressive work on fine-tuning IL policies with RL using sparse rewards!
@JiahengHu1
Jiaheng Hu
1 year
🚀 Despite efforts to scale up Behavior Cloning for Robots, large-scale BC has yet to live up to its promise. How can we break through the performance plateau? Introducing 🔥FLaRe: fine-tuning large-scale robot policies with Reinforcement Learning. https://t.co/iRC1NTgoFI 🧵
0
0
2