@Arjun_7m profile picture
Arjun Choudhry @Arjun_7m

1st Year ML PhD @GeorgiaTech. Previously: @AutonLab @SCSatCMU, @UTSAAII, @UQAM, @dtu_delhi. Interests: Multimodal FMs, Time Series, Efficiency

Joined July 2018
  • Tweets 115
  • Following 1,802
  • Followers 298
Photos and videos
Pinned Tweet
Arjun Choudhry@Arjun_7m
24 May 2025
🎓 Thrilled to announce I'll be joining @gtcomputing as an ML PhD student this fall, working with Prof. @badityap! Until then, continuing my research journey at @LTIatCMU with Professors @daphneipp and @MaartenSap. [1/2]
7
83
9,502
Arjun Choudhry retweeted
Duen Horng "Polo" Chau@PoloChau
21 Nov 2025
World's coolest #CSE school is hiring again! "AI and finance" is new this year!
Georgia Tech's School of Computational Science and Engineering is hiring faculty members across AI and finance, high-performance computing, data science and visual analytics, scientific computing and simulation, AI and machine learning, computational bioscience & biomedicine

ALT Georgia Tech's School of Computational Science and Engineering is hiring faculty members across AI and finance, high-performance computing, data science and visual analytics, scientific computing and simulation, AI and machine learning, computational bioscience & biomedicine

1
10
21
2,679
🎓 Thrilled to announce I'll be joining @gtcomputing as an ML PhD student this fall, working with Prof. @badityap! Until then, continuing my research journey at @LTIatCMU with Professors @daphneipp and @MaartenSap. [1/2]
7
83
9,502
Grateful to Dr. @MononitoGoswami, Prof. Artur Dubrawski, and the amazing @AutonLab team for their incredible support along the way. Excited for what's ahead! 🚀 [2/2]
8
800
Arjun Choudhry retweeted
Mononito Goswami
@MononitoGoswami
12 Feb 2025
🚀 Excited to share that MOMENT has been downloaded 1.3M times on HuggingFace, with over 280K downloads in just the last month! MOMENT is the BERT for time series. It can solve many tasks across different domains, including healthcare! Best part? It's completely open source!
3
5
22
1,199
Heartiest congratulations to our amazing team! I am applying for PhD positions for Fall 2025. If you see a fit with your lab or interests, or just want to brainstorm on ideas, please DM :)
Mononito Goswami
@MononitoGoswami
18 Dec 2024
Replying to @MononitoGoswami
Huge congratulations to my amazing collaborators Yifu Cai, @Arjun_7m, and Dr. Artur Dubrawski at @AutonLab @CMU_Robotics, @SCSatCMU
1
3
23
2,127
Excited to share TimeSeriesExam for systematic evaluation of time series reasoning capabilities of LLMs. Think your LLM can reason on time series concepts? Take it for a spin on the TimeSeriesExam! Now publicly available on HuggingFace :)
Mononito Goswami
@MononitoGoswami
18 Dec 2024
Frontier models like Gemini excel in areas like code and image understanding. But how well do they handle time series reasoning? (Re-)Introducing TimeSeriesExam, which just won the NeurIPS'24 TSALM Best Paper Honorable Mention🏆
6
13
878
Arjun Choudhry retweeted
Mononito Goswami
@MononitoGoswami
18 Dec 2024
Huge congratulations to my amazing collaborators Yifu Cai, @Arjun_7m, and Dr. Artur Dubrawski at @AutonLab @CMU_Robotics, @SCSatCMU
3
4
18
2,741
Arjun Choudhry retweeted
Mononito Goswami
@MononitoGoswami
18 Dec 2024
Frontier models like Gemini excel in areas like code and image understanding. But how well do they handle time series reasoning? (Re-)Introducing TimeSeriesExam, which just won the NeurIPS'24 TSALM Best Paper Honorable Mention🏆
Google DeepMind
@GoogleDeepMind
12 Dec 2024
The Gemini 2.0 era is here. And we’re excited for you to start building with it. A quick rewind of what we just released ⏪ Gemini 2.0 Flash ⚡ comes with low latency and better performance. 🔵 You can now access an experimental version in @GeminiApp on the web, while Gemini Advanced users can try Deep Research, a new AI research assistant. 🔵 Developers can begin building through the Gemini API in @Google AI Studio and Vertex AI 2.0 is also enabling new research prototypes of AI agents, including: 🔵 Project Astra, which explores future capabilities of a universal AI assistant 🔵 Project Mariner, which shows what’s possible for human-agent interaction, starting with your browser 🔵 Jules, an experimental AI-powered coding agent Finally, we’re exploring how 2.0 can be used in agents across domains — from navigating the virtual world of video games to applying its spatial reasoning capabilities to robotics. 🤖 goo.gle/gemini-2
0:14
2
7
25
5,829
Arjun Choudhry retweeted
Mononito Goswami
@MononitoGoswami
15 Dec 2024
Also my awesome collaborators and mentees, @Arjun_7m and @inverse_hessian are looking for PhD positions. If you are looking for students with experience in multimodal foundation models, efficiency, mechanistic interpretability, I don’t know of anyone better than them!
1
1
2
495
Arjun Choudhry retweeted
Mononito Goswami
@MononitoGoswami
26 Nov 2024
Replying to @zacharylipton
Super interesting! Even some recent work on clinical time series foundation models (trained on PPG / EEG), don't seem to be significantly better than general ones e.g. our model MOMENT
1
1
5
427
Arjun Choudhry retweeted
Mononito Goswami
@MononitoGoswami
31 Oct 2024
Time Series & Language: A Powerful Pair 📈📉 Loved seeing some recent work time series & text work! It reminded me of our very own work JoLT, one of the first time series and text models ! Want to learn more? Check out JoLT, and our latest paper, on the TimeSeriesExam!
An overview of JoLT. Given a time-series and an optional textual prompt as input, JoLT produces text as output. We pre-train a Transformer using the masked time-series reconstruction objective to use as an encoder, and the pre-trained language model as the decoder. The Q-Former is trained to align time-series and text representations. Learnable query tokens are used to extract time-series features conditioned on textual prompts.

ALT An overview of JoLT. Given a time-series and an optional textual prompt as input, JoLT produces text as output. We pre-train a Transformer using the masked time-series reconstruction objective to use as an encoder, and the pre-trained language model as the decoder. The Q-Former is trained to align time-series and text representations. Learnable query tokens are used to extract time-series features conditioned on textual prompts.

Each template in the TimeSeriesExam evaluates a specific category, and includes a question, list of options, example question and answer pair for in-context learning, and optionally a hint and descriptions of complicated technical terms. Here, GPT-4o showcases its ability to transfer visual understanding and time series concepts into effective reasoning.

ALT Each template in the TimeSeriesExam evaluates a specific category, and includes a question, list of options, example question and answer pair for in-context learning, and optionally a hint and descriptions of complicated technical terms. Here, GPT-4o showcases its ability to transfer visual understanding and time series concepts into effective reasoning.

1
13
35
3,658
Arjun Choudhry retweeted
Mononito Goswami
@MononitoGoswami
22 Jul 2024
🎊Excited to present our latest work on time series foundation models at #ICML2024 : ⌛️ MOMENT: A Family of Open Time-series Foundation Models, on Tuesday, 11:30 AM, Hall C 4-9 #305
Poster of the MOMENT Time Series Foundation Model poster.

ALT Poster of the MOMENT Time Series Foundation Model poster.

6
1
18
600
Arjun Choudhry retweeted
Konrad Szafer@KonradSzafer
5 Jul 2024
On the occasion of reaching 14k downloads and 200 stars for Moment 1, we are excited to release all our research code! The repository contains the necessary code to reproduce all experiments and models. github.com/moment-timeseries…
2
2
9
1,856
Arjun Choudhry retweeted
Mononito Goswami
@MononitoGoswami
18 Jun 2024
We loved building MOMENT-- one pre-trained model for many time series tasks. Have you tried it on your dataset yet? 📜: arxiv.org/abs/2402.03885 🧑‍💻: github.com/moment-timeseries… 🏋️: huggingface.co/AutonLab/MOME…
CMU Robotics Institute@CMU_Robotics
17 Jun 2024
The Auton Lab has proposed MOMENT: one of the first families of open-source foundation models for time series analysis! The models serve as a building block for diverse tasks such as forecasting, classification, & anomaly detection. Learn more: moment-timeseries-foundation…
6
15
1,980
Arjun Choudhry retweeted
Towards Data Science
@TDataScience
1 May 2024
MOMENT, a recently released family of foundation models, aims to tackle time series forecasting, classification, and more; @nikos_kafritsas explains how it works, unpacks its architecture, and compares its performance to other state-of-the-art models. buff.ly/3xRrKka
7
28
4,230
Arjun Choudhry retweeted
Konrad Szafer@KonradSzafer
10 May 2024
We are thrilled to announce that our paper "MOMENT: A Family of Open Time-series Foundation Models" has been accepted for the ICML 2024! On this occasion, we are open-sourcing it, together with the model weights and dataset! Links to everything in the thread! 🧵🤗
1
18
80
28,712
Arjun Choudhry retweeted
Mononito Goswami
@MononitoGoswami
10 May 2024
Exciting news! Just as foundation models revolutionized text and image modeling, it's now time for Time Series Data! Introducing MOMENT, a family of Open Time Series Foundation Models, freshly accepted at @icmlconf 2024! 🎉
4
15
68
12,473
Foundation models have revolutionized the way text and images are modeled. Now, it's the turn for time series data to experience their transformative impact. We are thrilled to introduce MOMENT, a family of open time series foundation models, recently accepted at ICML 2024. 1/n
1
7
12
2,110
more replies
For those interested in exploring MOMENT further, we provide the following resources: Website: moment-timeseries-foundation… Paper: arxiv.org/abs/2402.03885 Source Code: github.com/moment-timeseries… Pretrained Weights: huggingface.co/AutonLab/MOME… Training Dataset: huggingface.co/datasets/Auto… 3/n

1
143
We are thrilled to contribute to the advancement of time series modeling and look forward to the innovative applications and research that MOMENT will enable. Work done with @MononitoGoswami @KonradSzafer Yifu Cai, Shuo Li, and Prof. Artur Dubrawski. 4/n
101
Load more