Divyat Mahajan

@divyat09

Ph.D. Candidate @Mila_Quebec | Visiting Researcher @AIatMeta | Former: @MSFTResearch @IITKanpur

Montreal

divyat09.github.io

Joined August 2016

712 Following

830 Followers

Replies Reposts Media

Pinned
Divyat Mahajan
@divyat09
Oct 29, 2025
[1/9] While pretraining data might be hitting a wall, novel methods for modeling it are just getting started! We introduce future summary prediction (FSP), where the model predicts future sequence embeddings to reduce teacher forcing & shortcut learning. 📌Predict a learned
GIF
11
Divyat Mahajan
@divyat09
Oct 21
Exciting work led by @hazel_heejeong on benchmarks for evaluating compositional generalization in world models!
Hazel Nam
@hazel_heejeong
Oct 21
Your world model may know less about the world than you think. I've always wondered what world models really learn. Humans don't memorize every trajectory. We learn rules we can reuse. Can world models do the same? We built AnyGoal to find out. hazel-heejeong-nam.github.io/anygoal/
Divyat Mahajan
@divyat09
Sep 15
Really exciting to see autoresearch agents for causal inference, with formal verification in the loop!
Vasilis Syrgkanis
@syrgkanis
Sep 14
Interesting times! Thanks to the amazing work of my student @JiyuanTan, we recently released a fully automated agentic pipeline that automates theoretical research in causal inference jiyuan-tan.github.io/CausalSmith/.
Divyat Mahajan
@divyat09
Apr 21
Presenting Future Summary Prediction at #ICLR2026 🇧🇷 📌 Friday, 24th April, 10.30 am 📍Pavilion 3 (521) Come over to chat about novel pretraining objectives for LLMs!
Divyat Mahajan
@divyat09
Oct 29, 2025
[1/9] While pretraining data might be hitting a wall, novel methods for modeling it are just getting started! We introduce future summary prediction (FSP), where the model predicts future sequence embeddings to reduce teacher forcing & shortcut learning. 📌Predict a learned
1
Divyat Mahajan
@divyat09
Apr 20
[1/8] In-context learning and prior-fitted networks are gaining traction for causal effect estimation in tabular datasets. But can they also generate novel observational & interventional data from causal models? 📌Introducing Cond-FiP: In-context learning of causal mechanisms
2

Divyat Mahajan

Divyat Mahajan’s posts