Amartya Dutta

37 posts

Amartya Dutta banner
Amartya Dutta

Amartya Dutta

@amadutx

PhD Student @VT_CS , @SanghaniCtrVT | BTech @IIITGhy | Computer Vision, LLMs, MLLMs

Blacksburg, VA Katılım Ekim 2020
214 Takip Edilen42 Takipçiler
Amartya Dutta
Amartya Dutta@amadutx·
🎉 Excited to share that "Open World Scene Graph Generation using Vision Language Models" has been accepted at #ICML2025 as a Workshop Paper! It’ll be presented at the @WorldModelsICML Workshop. 🗓️July 18, 3:45–5:00 pm PST. Feel free to reach out if you're interested in our work!
The Sanghani Center at Virginia Tech@SanghaniCtrVT

Open World Scene Graph Generation using Vision Language Models Workshop @amadutx @ksmehrab @medha_sawhney @neog_abhilashh @mridulk97 @SepidehFatemi A. Pradhan (all VT) M. Maruf @Ismini_L @arkadaw_ @anujkarpatne (VT) arxiv.org/abs/2506.08189

English
2
3
8
619
Amartya Dutta retweetledi
Jinay
Jinay@jinaycodes·
Introducing soarXiv ✈️, the most beautiful way to explore human knowledge Take any paper's URL and replace arxiv with soarxiv (show in video) to teleport to its place in the universe I've embedded all 2.8M papers up until April 2025 Try it at: soarxiv dot org
English
151
1.1K
9.1K
513.3K
Amartya Dutta retweetledi
Medha Sawhney
Medha Sawhney@medha_sawhney·
Do swing by our #ICLR2025 poster on jointly solving forward & inverse problems in geophysics! 🌍 Presented by @AnujKarpatne! 🗓️ Sat, Apr 26 | 10:00–12:30 +08 📍 Poster #195 | Hall 3 + Hall 2B 🔗kgml-lab.github.io/projects/GFI-f… I won’t be there😞, but glad to chat anytime!! #AI4Science
Medha Sawhney tweet media
Medha Sawhney@medha_sawhney

🌟Thrilled to share our paper, “A Unified Framework for Forward and Inverse Problems in Subsurface Imaging using Latent Space Translations”, has been accepted at #ICLR2025 ! 🎉 Grateful to my co-authors, Naveen Gupta, @arkadaw_ , @YouzuoLin1 and @anujkarpatne, for their support.

English
0
5
29
2K
Amartya Dutta retweetledi
Medha Sawhney
Medha Sawhney@medha_sawhney·
🌟Thrilled to share our paper, “A Unified Framework for Forward and Inverse Problems in Subsurface Imaging using Latent Space Translations”, has been accepted at #ICLR2025 ! 🎉 Grateful to my co-authors, Naveen Gupta, @arkadaw_ , @YouzuoLin1 and @anujkarpatne, for their support.
Anuj Karpatne@anujkarpatne

Happy to share 2 papers from #KGML lab got accepted to #ICLR2025! Kudos to my students Harish, Naveen, and @medha_sawhney for leading these works (their first ICLR papers!) and to our amazing set of collaborators! @imageomics @SanghaniCtrVT @YouzuoLin1

English
1
6
14
2.9K
Amartya Dutta retweetledi
Omar Sanseviero
Omar Sanseviero@osanseviero·
BREAKING NEWS The Royal Swedish Academy of Sciences has decided to award the 2024 #NobelPrize in Literature to the Attention Is All You Need authors. Their work has made thousands cry, laugh, or rich and made GPUs go brrr
Omar Sanseviero tweet mediaOmar Sanseviero tweet media
English
31
170
2K
161.7K
Amartya Dutta retweetledi
Anuj Karpatne
Anuj Karpatne@anujkarpatne·
@ulrichpaquet @NicLoizou @jmtomczak @DaniCMBelg We are facing a similar issue with our #NeurIPS2024 submission (ID 19132) with review scores of 5 7 7 6 (avg score 6.25) being rejected by the meta-reviewer on an unreasonable concern not mentioned in any of the reviews. I just emailed pc2024@neurips.cc. Hoping it gets looked.
English
1
9
11
957
Amartya Dutta retweetledi
Yu Su
Yu Su@ysu_nlp·
Introducing BioCLIP: A Vision Foundation Model for the Tree of Life imageomics.github.io/bioclip/ A foundation model that strongly generalizes on the tree of life (2M+ species), outperforming OpenAI CLIP by 18% in zero-shot classification, and supports open-ended classification over almost the entire tree of life What's the secrete ingredients? > Data: we curate and release TreeOfLife-10M, the largest and most diverse ML-ready dataset of organism images to date. It contains 10.4M images for over 450K taxa, sourced from iNaturalist, BIOSCAN, and Encyclopedia of Life. > Modeling: we creatively repurposes CLIP's multimodal contrastive learning objective for hierarchical image classification. The autoregressive language model naturally encodes the hierarchy of the tree of life taxonomy, which in turn bakes the hierarchical representation into the vision transformer encoder. Key results > Strong zero/few-shot classification for animals/plants/fungi, including rare species, outperforming CLIP by avg 16-18% absolute. > T-sne visualization shows that BioCLIP's vision encoder has captued the fine-grained hierarchical structure of the tree of life > BioCLIP is a kind of universal classifier for the tree of life. Just give it an organism image and it will likely find the correct species (among top 5)! But use it with caution; it's not perfect yet.. Final remarks > AI for Science is really hard but extremely rewarding! It took us a ton of time (1+ year) and frustration trying to find a plausible way to integrate the tree of life taxonomy into foundation model training. But when the "Eureka!" moment came and the idea hit us (by the great @weilunchao) that CLIP's multimodal contrastive learning objective can be repurposed for that, everything just follows naturally. It was truly a moment of joy and excitement! > BioCLIP is our first attempt at foundation models for biology, but it certainly won't be the last! There's so much more to do at the intersection of one of the oldest scientific disciplines and the young but thriving field of AI. Biological intelligence is the foundation for artificial intelligence, and artificial intelligence will in turn become the most important tool for us to unraval the mysteries of biological intelligence. We are hiring postdocs and PhDs in the NSF @imageomics institute to explore this exciting field! Drop us an email. also happy to chat about it at #NeurIPS2023 with any of Tanya, @weilunchao, or me. - paper: arxiv.org/abs/2311.18803 - project: imageomics.github.io/bioclip/ - demo: huggingface.co/spaces/imageom… - model: huggingface.co/imageomics/bio… - data (TreeOfLife-10M): to be released on Hugging Face soon joint work with the amazing @imageomics team: @samstevens6860 Lisa Wu, Matt Thompson, Elizabeth Campolongo @luke_ch_song @Carlyn2015 @donglixp @dahdulw Chuck Stewart, Tanya Berger-Wolf @weilunchao @ysu_nlp
English
9
90
428
80.6K
Arka Daw
Arka Daw@arkadaw_·
Happy to share that I've been selected as a Top reviewer for #NeurIPS2023 and have received a complimentary registration! A huge shoutout to the organizers and area chairs for this honor. 🥳 Link: #top-reivewers" target="_blank" rel="nofollow noopener">neurips.cc/Conferences/20…
English
4
0
15
1.1K
Amartya Dutta retweetledi
MatLab crashes
MatLab crashes@memecrashes·
MatLab crashes tweet media
ZXX
18
423
3.9K
221.8K
Amartya Dutta retweetledi
Anuj Karpatne
Anuj Karpatne@anujkarpatne·
Curious to learn how images in biology can be grounded with scientific knowledge? @ElhamodMohannad will be presenting the first @imageomics paper at KDD 2023 on Phylogeny-guided NNs. If you are at KDD, stop by at 2 pm on Thursday in 102A to learn more! @kdd_news @SanghaniCtrVT
English
0
7
12
1.3K
Shivangi Srivastava
Shivangi Srivastava@shivangi_sriv·
Looking for a video editor to edit a high production quality short videos. P.S. this is a long term engagement for a passion project
English
9
5
11
5.2K
ESPN FC
ESPN FC@ESPNFC·
PSG AND RIYADH ALL-STAR'S PUT ON A SHOW IN SAUDI ARABIA ✨
ESPN FC tweet media
English
131
831
10.6K
487.6K
Rajat Kanti Bhattacharjee
Rajat Kanti Bhattacharjee@simpuleguy·
About to take my 1st interview and I think I am more nervous than the guy who will be giving it. 😶 🥲 He he he meko good luck bolo koi 🤣
English
1
0
1
0
Amartya Dutta retweetledi
Jai Prakash Yadav
Jai Prakash Yadav@Jaiprakash11_JP·
Help save Aman!🙏🏻 Aman, a third year undergrad at IIIT Guwahati is in a life-and-death battle with a sudden but devastating lung infection. Perhaps the kindest and the most jovial person in his class, right now he needs your help to survive.(1/4)
English
98
229
216
0