Sarath Chandar's research group at @polymtl, @UMontreal and @Mila_Quebec focusing on Machine Learning!
Montréal, Québec, Canada
- Tweets 181
- Following 146
- Followers 734
- Likes 110
🎓 Graduate Program Application Mentorship
Applying to grad school doesn't have to be overwhelming!
Get personalized 1:1 support from students who’ve successfully navigated the process.
🔗 Apply now: forms.gle/FyjEuLmjcdREF3Ba6
🗓️ Deadline: Nov 1, 2026
👇 Here is what you get:
1
14
52
12,377
✨ What you’ll get from our mentorship program:
📄 CV & SOP Feedback to polish your documents
🎙️ Interview Tips to build real confidence
🤝 1:1 Guidance directly from current grad students
1
722
Don't navigate the complex application process alone. We're here to help bridge the gap and showcase your full potential! 🌟
#GradSchool #Mentorship #TechDiversity #HigherEd #MachineLearning #EDI #AI
486
🎉 Thrilled to share that 8 papers from our lab (and collaborations with lab members) have been accepted at #NeurIPS2026!
The work spans RL, world models, dynamical systems, LLM safety & interpretability, tabular foundation models, and forecasting. Huge congrats to everyone 🧵👇
2
20
61
4,583
📄 φTD: Distributional Reinforcement Learning using Characteristic Functions
Distributional RL built on characteristic functions.
w/ @DavideBald42296
1
2
166
Congrats again to all the students and collaborators behind these papers! 🙌 See you at #NeurIPS2026
1
141
Chandar Lab retweeted
Just got accepted at NeurIPS! Read our paper on how (we think) TFMs generalize!
arxiv.org/pdf/2608.17957
The Bayesian interpretation is insufficient for Tabular Foundation Models.
A model trained on MNIST generalizes in-context to California Housing, even though the prior places no support on the downstream task.
In this new pre-print, we try to explain why.
(arXiv: 2608.17957)
3
2
27
1,649
Chandar Lab retweeted
Our semantic world models paper is accepted at #NeurIPS2026 E&D Track! 🎉
What's the best latent space for diffusion world models? We find semantic spaces beat VAE-style ones on planning and robustness, with competitive visual quality.
Paper & more: hskalin.github.io/semantic-w…
Diffusion world models can help test and improve robot policies before running them on real robots.
But can the choice of latent space make the WM more faithful?
We show that semantic spaces beat reconstruction spaces on task relevant metrics.
hskalin.github.io/semantic-w…
3
4
30
2,042
Chandar Lab retweeted
Read our paper on scaling interpretable LLMs, we show that interpretable LLMs are not only possible but can scale both on typical generation benchmarks and interpretability benchmarks. arxiv.org/abs/2608.07594👀
4
20
139
17,108
Chandar Lab retweeted
People use LLMs all the time for help with medical problems, some doctors might also use them too to explain diagnoses to patients.
But diagnostic tests are inherently probabilistic. This work from @dicemasz explores if LLMs are suited or not for this usecase!
Turns out that many contemporary LLMs still struggle to explain probabilities and uncertainty in plain natural language 👀
1
2
151
Turns out that many contemporary LLMs still struggle to explain probabilities and uncertainty in plain natural language 👀
1
12
14
3,412
Such findings call LLMs’ reliability as explainers of probability into question — they may not be well-suited to explaining other predictive outputs to decision-makers or discussing future uncertainties in a calibrated manner.
1
1
110
We got these & other insights from @dicemasz main MSc work, which focused on evaluating LLMs’ limitations for communicating risk in natural language — already out on preprint if you want to learn more!!
See: arxiv.org/abs/2607.03882
3
87