JACQUELINE R. M. A. MAASCH

           



I am a fifth-year PhD candidate in computer science at Cornell Tech and the Weill Cornell Medicine Institute of AI for Digital Health. My doctoral research has been supported by the NSF Graduate Research Fellowship, Cornell's Presidential Life Science Fellowship, and the Digital Life Initiative. I am advised by Dr. Fei Wang, Dr. Volodymyr Kuleshov, and Dr. Kyra Gan. Previously, I was a research intern at Microsoft Research Cambridge.

I am interested in machine learning for reasoning and decision-making under uncertainty. This includes open problems in AI reasoning: building reasoning machines, what that requires in theory and practice, and its consequences for society. I approach these problems primarily through the theoretical frameworks of probabilistic and causal graphical modeling.

Jacqueline Maasch headshot

NEWS


I will be entering the job market in 2026. If you are interested in collaborating or would like to schedule a meeting at NeurIPS 2025, please reach out.

RECENT WORK


A subset of my lead-author, peer-reviewed papers are highlighted below. My full bibliography is on Google Scholar.

CausalARC: Abstract Reasoning with Causal World Models. J Maasch, J Kalantari, K Khezeli.

[tl;dr] An experimental testbed for reasoning under distribution shift with few-shot learning and test-time adaptation.

NeurIPS LAW 2025 (spotlight)  ✦︎  arxiv  website  🤗 hf

Probabilistic Graphical Models: A Concise Tutorial. J Maasch, W Neiswanger, S Ermon, V Kuleshov.

[tl;dr] This 200-page tutorial reviews the theory and methods of representation, learning, and inference in probabilistic graphical modeling.

under review  ✦︎  arxiv  website

Compositional Causal Reasoning Evaluation in Language Models. J Maasch, A Hüyük, X Xu, A Nori, J González.

[tl;dr] A novel evaluation framework reveals taxonomically distinct error patterns when LLMs reason over the composition of causal measures.

ICML 2025  ✦︎  arxiv  website  code  poster  slides  🤗 hf - 2k+ downloads

Local Causal Discovery for Structural Evidence of Direct Discrimination. J Maasch, K Gan, V Chen, A Orfanoudaki, N Akpinar, F Wang.

[tl;dr] Local structure learning facilitates causal fairness analysis in complex decision systems, as illustrated by a real-world case study on organ transplant allocation.

AAAI 2025  ✦︎  arXiv  code  poster  slides

Local Discovery by Partitioning: Polynomial-Time Causal Discovery Around Exposure-Outcome Pairs. J Maasch, W Pan, S Gupta, V Kuleshov, K Gan, F Wang.

[tl;dr] Local causal discovery enables efficient automated covariate selection under limited prior knowledge with guarantees on correctness.

UAI 2024  ✦︎  arXiv  code  poster  slides

Molecular de-extinction of ancient antimicrobial peptides enabled by machine learning. J Maasch*, M Torres*, M Melo, C de la Fuente.

[tl;dr] Computational proteolysis guides bioinspired prospection for encrypted antimicrobial peptides that display in vitro and in vivo activity with low host toxicity.

Cell Host & Microbe 2023  ✦︎  paper  code  npr  nature news  cnn  vox

CONTACT

Preferred contact: In general, I can be reached through LinkedIn.

Pronouns: they or she.
If you are new to nonbinary pronouns, here are some examples for how to use them in a grammatical way from Merriam Webster, the MLA Style Guide, and the APA Style Guide. ☺ 

echo @ | sed 's/^/maasch/' | sed 's/$/cs.cornell.edu/'

Updated November 2025. This website was adapted by J. Maasch from this source code.