Print this page to PDF to save a copy. The content is editable for light modifications. Elements with hide-on-print (like this one) won't show. Use browser print settings to customize the output till relevant sections fit on one page. Feel free to copy this style (Right Click > View Page Source)!

Atharva Sehgal

[email protected]
atharvas.net
linkedin.com/in/atharvas

Research Interests

My research focuses on developing self-improving AI algorithms for automated and collaborative scientific discovery across scientific domains and modalities.

Education

Postdoctoral Researcher, Caltech
Working with Yisong Yue on building self-evolving AI agents.
June 2026 - Present
PhD in Computer Science, University of Texas Austin
Developed code generation algorithms for world modelling, visual reasoning, and scientific discovery. Advised by Swarat Chaudhuri.
August 2021 - May 2026
B.S. in Computer Science, University of Illinois Urbana Champaign
Graduated with high honors. Minor in Linguistics. James Scholar.
August 2017 - May 2021

Publications

Programmatic Context Augmentation for LLM-based Symbolic Regression Preprint 2026
Hao Liu, Xiao-Wen Yang, Atharva Sehgal, Yixin Wang, Lan-Zhe Guo, Yu-Feng Li, Yisong Yue
https://arxiv.org/abs/2605.03101
FormulaCode: Evaluating Agentic Optimization on Large Codebases ICML 2026
Atharva Sehgal*, James Hou*, Akanksha Sarkar, Ishaan Mantripragada, Swarat Chaudhuri, Jennifer J. Sun, Yisong Yue
https://formula-code.github.io/
VendiEvolve: Towards Diversity-Aware LLM-Guided Evolutionary Search GenAICreativity @ ICML 2026
Lei Lei, Atharva Sehgal, Lauren Hyoseo Yoon, Xuefei Wang, Swarat Chaudhuri, Yisong Yue
PWW-Bench: Probing Visual Mathematical Reasoning with Proofs Without Words AI4Math @ ICML 2026
Rahul Saha, Sabrina Reguyal, Alan Li, Atharva Sehgal
Simple Agents Outperform Experts in Biomedical Imaging Workflow Optimization CVPR 2026
Xuefei (Julie) Wang, Kai A. Horstmann, Ethan Lin, Jonathan Chen, Alexander R. Farhang, Sophia Stiles, Atharva Sehgal, Jonathan Light, David Van Valen, Yisong Yue, Jennifer J. Sun
https://arxiv.org/abs/2512.06006
Beyond Accuracy: Metrics that Uncover What Makes a 'Good' Visual Descriptor (Best Poster) | VisCon @ CVPR 2025
Ethan Lin, Linxi Zhao, Atharva Sehgal, Jennifer J. Sun
https://arxiv.org/abs/2507.03542
Escher: Self-Evolving Visual Concept Library using Vision-Language Critics CVPR 2025
Atharva Sehgal, Patrick Yuan, Ziniu Hu, Yisong Yue, Jennifer J. Sun, Swarat Chaudhuri
https://arxiv.org/abs/2504.00185
LaSR: Symbolic Regression with a Learned Concept Library NeurIPS 2024
Arya Grayeli*, Atharva Sehgal*, Omar Costilla-Reyes, Miles Cranmer, Swarat Chaudhuri
https://arxiv.org/abs/2409.09359
Neurosymbolic Grounding for Compositional World Models ICLR 2024
Atharva Sehgal, Arya Grayeli, Jennifer J. Sun, Swarat Chaudhuri
https://arxiv.org/abs/2310.12690
Neurosymbolic Programming for Science AI4Science @ NeurIPS 2022
Jennifer J. Sun*, Megan Tjandrasuwita*, Atharva Sehgal*, Armando Solar-Lezama, Swarat Chaudhuri, Yisong Yue, Omar Costilla-Reyes
https://arxiv.org/abs/2210.05050
Composing Neural and Symbolic Reasoning with an Application to Visual Discrimination IJCAI/ECAI 2022
Adithya Murali, Atharva Sehgal, Paul Krogmeier, P. Madhusudan
https://arxiv.org/abs/1907.05878
Statheros: A Compiler for Efficient Low-Precision Probabilistic Programming DAC 2021
Jacob Laurel, Rem Yang, Atharva Sehgal, Shubham Ugare, Sasa Misailovic
https://ieeexplore.ieee.org/document/9586276

Projects

tacc-inference Library August 2024
The tacc-inference library provides a common API to run LLMs on a single node or multiple nodes of TACC's Vista and Frontera supercomputers (the largest academic supercomputer in the US). Used by 50+ labs. (pip install tacc-inf)
neurosym Library August 2023
The neurosym library is a Python package for neuro-symbolic program synthesis. It is the first framework which integrates tools for DSL design, program search, and program abstraction in a self contained package. Used extensively in research, production, and teaching. Joint work with collaborators at MIT. (pip install neurosym)

Technical Strengths

Computer Languages: Python, C, Julia, C++14, Haskell, HTML/CSS/JavaScript, OCaml
Frameworks: PyTorch/TensorFlow/Scipy, Pandas/Dask, NetworkX, Coq/Lean, Z3, Pyro

Outreach, Service, and Talks

Academic Reviewing
ICML (2023-present), NeurIPS (2022-present), ICLR (2023-present) CVPR (2025)
Invited Talks
Can AI Agents holistically optimize codebases? Cornell (2026)
Can AI Agents holistically optimize codebases? Caltech (2026)
Program Synthesis and Scientific Discovery MIT (2025)
Program Synthesis and Scientific Discovery Caltech (2024)
Program Synthesis and Scientific Discovery Cornell (2024)
Neurosymbolic Programming and Scientific Discovery Chalmers University (2024)
Tutorial on Neurosymbolic Programming Caltech (2022)
Tutorial on Neurosymbolic Programming POPL (2023)
Tutorial on Neurosymbolic Programming ICSE (2024)
Tutorial on Neurosymbolic Programming MIT (2024)
Teaching
Math Tutor Caltech Y (FA24, SP25)
College Math Prep (Co-Instructor) Coleman State Prison (FA23)
DiRP: Neurosymbolic Programming (Instructor) UT Austin (SP24, FA23, SP23)
Honors: Embedded Systems (TA) UIUC (SP20)
Honors: Algorithms for String Processing (TA) UIUC (FA20, SP21)
Data Structures and Algorithms (TA) UIUC (FA19, SP20, FA20, SP21)
Discrete Mathematics (TA) UIUC (FA20)