Shweta Mahajan

162 posts

Shweta Mahajan

Shweta Mahajan

@Matewhs

Assistant Professor at York University || Vector Institute Postdoc, University of British Columbia || Ph.D., TU Darmstadt

Toronto, ON Katılım Haziran 2020
220 Takip Edilen188 Takipçiler
Xindi Wu
Xindi Wu@cindy_x_wu·
So honored to receive the outstanding paper honorable mention for our motion attribution for video generation paper at #ICML2026 This wouldn’t happen without my amazing mentors and advisors! @paschalidoud_1 @JunGao33210520 Antonio Torralba @lealtaixe @orussakovsky @FidlerSanja @jonLorraine9 and many other friends and colleagues, this work would not have been possible without all of you and the huge support from @NVIDIAAI @PrincetonCS @MIT_CSAIL @UMichCSE @UofTCompSci @VectorInst 💪
ICML Conference@icmlconf

🏆Announcing the #ICML2026 Awards! 🏆 Including Outstanding Papers (research paper & position paper, winner & honorable mentions) and the Test of Time Award! Check out the blog post for all winners (or read on), laudatio, & description of the processes. blog.icml.cc/?p=1347

English
17
13
205
15.7K
Shweta Mahajan
Shweta Mahajan@Matewhs·
Our new benchmark and dataset for interactive agents. The AI assistants are expected to interrupt when a mistake is made and provide corrective feedback! The benchmark is high-quality collected by our own research team! @apratimbh @RolandMemisevic
Apratim Bhattacharyya@apratimbh

🚨Introducing: Ego-MC-Bench (Mistake Corrections) benchmark and Ego-CoMist (Counterfactual Mistakes) dataset. 🎯Ego-MC-Bench: Where AI assistants need to intervene at the right time (when) and with the right feedback (what) to prevent mistakes. 👉tinyurl.com/y7y9mwrs 1/4

English
0
0
2
109
Shweta Mahajan retweetledi
Apratim Bhattacharyya
Apratim Bhattacharyya@apratimbh·
🚨Introducing: Ego-MC-Bench (Mistake Corrections) benchmark and Ego-CoMist (Counterfactual Mistakes) dataset. 🎯Ego-MC-Bench: Where AI assistants need to intervene at the right time (when) and with the right feedback (what) to prevent mistakes. 👉tinyurl.com/y7y9mwrs 1/4
English
1
6
15
11.7K
Anand Bhattad
Anand Bhattad@anand_bhattad·
[5/5] Camera conditioning improves performance across ACT, Diffusion Policy, and SmolVLA, in both simulation and the real world. Key takeaway: Robot policies may spend substantial capacity inferring camera pose from appearance cues. When camera pose is available (or can be estimated), providing it explicitly can significantly improve viewpoint generalization. Full video below 👇 Paper: arxiv.org/abs/2510.02268 Project: ripl.github.io/know_your_came… code: github.com/ripl/CamPoseOp… Congratulations to lead author @JiangTianchong and the entire team: Jingtian Ji, Xiangshan Tan, @jiading_fang, @vitorguizilini, @mattrwalter
English
1
1
2
451
Anand Bhattad
Anand Bhattad@anand_bhattad·
[1/5] 🏆 Honored that our paper received the #ICRA2026 Best Paper Award on Robot Learning. A simple question motivated this work: Do robot policies really understand viewpoint changes, or are they exploiting visual shortcuts hidden in the data? Modern robot policies are often evaluated across different camera viewpoints. When they succeed, we often conclude that they have learned viewpoint-invariant representations. Our findings suggest that this conclusion can be misleading.
Anand Bhattad tweet mediaAnand Bhattad tweet media
English
2
14
108
7.2K
Princeton Computer Science
Princeton Computer Science@PrincetonCS·
.@adjiboussodieng has won a CAREER award from the @NSF, a top honor for early career faculty! Dieng is working to establish a new paradigm for AI, transforming it into an engine for scientific exploration that can uncover revolutionary new materials. bit.ly/4udRPRK
Princeton Computer Science tweet media
English
1
18
42
24.8K
Shweta Mahajan retweetledi
Adi Chinchure
Adi Chinchure@adityachinchure·
Today’s the day! Come attend our workshop on cognitive-inspired VLMs, starting at 1 PM #CVPR2026
Adi Chinchure tweet media
English
0
4
7
1.4K
William Yang
William Yang@YangWilliam_·
I will be presenting BeyondObjects this week at #CVPR2026. Come check out our poster on Thursday 4:30-5:30pm at the syndata4cv workshop (Room 607) or Saturday 4:45-6:45pm at ExHall A & F poster location 99. Happy to chat!
William Yang tweet media
William Yang@YangWilliam_

Text-to-image (T2I) models can generate rich supervision for visual learning but generating subtle distinctions still remains challenging. Fine-tuning helps, but too much tuning → overfitting and loss of diversity. How do we preserve fidelity without sacrificing diversity (1/8)

English
1
5
17
3K
Shweta Mahajan retweetledi
Sanjay Haresh
Sanjay Haresh@SanjayHaresh·
📢Excited to be presenting our work on memory + VLAs at ICRA'26 this Thursday morning (poster 224). We found that a super simple language-based scratchpad with spatial and temporal grounding goes a long way in imparting memory to VLAs. 1/n
English
1
5
19
1.6K
Shweta Mahajan retweetledi
Apratim Bhattacharyya
Apratim Bhattacharyya@apratimbh·
📢Current VLAs are stateless - this is not ideal for long horizon tasks - which requires models to remember key past information, e.g., positions, orientations, task progress. 🎯So we just let the VLA take notes to help remember the past a.k.a "Notes-to-Self"! 👉ICRA 2026
Sanjay Haresh@SanjayHaresh

📢Excited to be presenting our work on memory + VLAs at ICRA'26 this Thursday morning (poster 224). We found that a super simple language-based scratchpad with spatial and temporal grounding goes a long way in imparting memory to VLAs. 1/n

English
0
2
3
268
Shweta Mahajan retweetledi
CogVL @ CVPR 2026
CogVL @ CVPR 2026@CogVLWorkshop·
CogVL Workshop at #CVPR2026 is less than a week away! We have an exciting lineup of keynote speakers across Vision, NLP and Cog Sci, and orals/posters on reasoning methods for VLM models. 🕐 June 3, 1 PM 📍 Rooms 610/612 🔗 Schedule: cogvl.github.io
CogVL @ CVPR 2026 tweet media
English
1
3
4
983

Keşfet

@cindy_x_wu @paschalidoud_1 @JunGao33210520 @lealtaixe @orussakovsky @FidlerSanja @jonLorraine9 @NVIDIAAI