@SachdevaBhuvan

MTS at @stellonlabs | Past Research at @MSFTResearch and @AmazonScience

bhuvan-21.github.io/
Joined December 2019
  • Tweets53
  • Following955
  • Followers118
3Media
Our paper has been accepted for Oral presentation at #CVPR 🎉🎉 Kudos to the team: @karan_uppal3, @abhinav_java See you in Denver! On a side note, I am finishing my fellowship by June and looking for full-time research roles. DMs open.
Bhuvan Sachdeva@SachdevaBhuvan6 Dec 2025
New paper: Understanding Task Transfer in Vision–Language Models How does finetuning a model on one task affect its performance on other tasks? @karan_uppal3 and @abhinav_java are presenting this work at Unireps, NeurIPS!! 📍 Ballroom 20D ⏰ 3:45 PM – 5:00 PM Come and say Hi!🧵
4
5
2
31
3,717
Can you run GTA VI on it though?
World Labs
@theworldlabsSep 1
Introducing Atlas: The world's first multimodal world model that generates image and video frames with pixel-perfect camera control and reconstructs them in 3D. Model the world, move the camera, and simulate space & time.
67
Belated update: I wrapped up my fellowship at Microsoft Research India last month and recently joined @stellon_labs as a Member of Technical Staff. I'm excited to be working on a completely new set of challenges with @ron_joshi and @divamgupta.
1
18
1,072
A very thorough and refreshing take on AI research.
Jack Morris
@jxmnopJun 15

Zen and the Art of AI Research

Article · @jxmnop
1
97
Bhuvan Sachdeva retweeted
Karan Uppal@karan_uppal3May 31
Excited to be heading to #CVPR2026 to give an oral presentation of our paper on task transfer! If you're in Denver, do attend the presentation on Sunday. Looking forward to meeting fellow researchers! Link: aka.ms/task-transfer-vlms Special thanks: @SachdevaBhuvan @abhinav_java
1
9
576
Checkout our latest work!!
Deepti Ghadiyaram@deeptigpMay 24
[1/4] The human eye doesn't process every single pixel of a video continuously—it focuses on what changes. So why are our video AI models wasting compute on redundant frames? Introducing Swift Sampling: a test-time technique inspired by the human visual system. 🧠👇
6
160
Checkout the exciting work by Ishaan!
Ishaan Watts@IshaanWatts18May 8
Spending billions to train the "best" base model? You might be optimizing the wrong thing! 🎯 We show that controlling sharpness during mid-training leads to over 35% less forgetting after fine-tuning / quantization... even when the base model itself gets worse. 🧵 Takeaways for pretraining: - Use SAM (Sharpness-Aware-Minimization) in the final steps (~10%) - Try much higher learning rates (yes, even ~10× larger) 1/9
2
211
Bhuvan Sachdeva retweeted
Saining Xie
@sainingxieApr 23
vision🍌 is here vision-banana.github.io/ if you got into computer vision the way I did, starting with pixel-level labeling tasks like segmentation, edges, depth, or surface normals, you’ll probably feel the same seeing these results -- something big has quietly shifted, and it’s going to change how we approach these problems for good 🧵
11
110
16
788
68,424
Excited to announce that I will be giving an oral presentation of BEAVER at the VerifAI workshop @ ICLR 2026! Come chat if you are interested in formal guarantees for LLMs 🦫
Nalin Wadhwa @ ICLR 2026@nalin_wadhwaFeb 5
🚨 Can we truly prove that LLMs won't leak your data or generate unsafe code? Almost all LLM safety evals today only give estimates. Sampling. Benchmarks. Leaderboards. Zero guarantees ❌ We present BEAVER, the first system to provably bound LLM constraint satisfaction. 🧵👇
1
4
18
1,866
Bhuvan Sachdeva retweeted
André Araujo@andrefaraujoApr 15
True multimodal AI needs to understand the world spatially 🎯 🚀 Excited to release #CVPR2026 TIPSv2 from @GoogleDeepMind, a foundational image-text encoder with spatial awareness, leading to strong overall results and massive gains on patch-text alignment. 🔥 1/N
11
96
9
732
84,620
Made a public RL-for-LLMs reading list because I was trying to prepare for my interviews 96 papers, 5 categories, 24 subtopics, mostly around the 2025-2026 wave, with notes on what’s worth reading carefully vs what you can skim. Hopefully useful if you’re getting into RLHF/agentic RL/ reward modeling, or just cramming for interviews. algoroxyolo.github.io/blog/2… Note: Paper selection mainly adhere to me and @sun_hanchi's taste... Make sure you follow me so I can have the incentive to update this and the agentic system design series #RLHF #LLM #AIAgents
18
89
2
936
56,117
Pretty good read, it reminded me of "You and Your Research" by Richard Hamming (which also referred in the article itself).
Nathan Lambert
@natolambertMar 10
Good post on how to think about honing your skills as an (academic) researcher by Carlini nicholas.carlini.com/writing…
1
136
Full paper accepted to CVPR!!😀
Bhuvan Sachdeva@SachdevaBhuvan6 Dec 2025
New paper: Understanding Task Transfer in Vision–Language Models How does finetuning a model on one task affect its performance on other tasks? @karan_uppal3 and @abhinav_java are presenting this work at Unireps, NeurIPS!! 📍 Ballroom 20D ⏰ 3:45 PM – 5:00 PM Come and say Hi!🧵
7
4
1
83
6,727
Checkout Nalin's first PhD project!!
Nalin Wadhwa @ ICLR 2026@nalin_wadhwaFeb 5
🚨 Can we truly prove that LLMs won't leak your data or generate unsafe code? Almost all LLM safety evals today only give estimates. Sampling. Benchmarks. Leaderboards. Zero guarantees ❌ We present BEAVER, the first system to provably bound LLM constraint satisfaction. 🧵👇
1
1
478
AI generated references about to go insane!
OpenAI
@OpenAIJan 27
Introducing Prism, a free workspace for scientists to write and collaborate on research, powered by GPT-5.2. Available today to anyone with a ChatGPT personal account: prism.openai.com/
2
215
New paper: Understanding Task Transfer in Vision–Language Models How does finetuning a model on one task affect its performance on other tasks? @karan_uppal3 and @abhinav_java are presenting this work at Unireps, NeurIPS!! 📍 Ballroom 20D ⏰ 3:45 PM – 5:00 PM Come and say Hi!🧵
1
5
1
16
10,116
more replies
(4/5) Why use PGF? PGF-guided selection enables alternative datasets that rival or surpass direct finetuning when supervision data is scarce.
1
1
279
(5/5) For more details, check out the paper: arxiv.org/abs/2511.18787. Happy to answer questions! Shoutout to my amazing co-authors and mentor Vineeth N B.

Understanding Task Transfer in Vision-Language Models

Vision-Language Models (VLMs) perform well on multimodal benchmarks but lag behind humans and specialized models on visual perception tasks like depth estimation or object counting. Finetuning on...

arxiv.org
3
181
Load more