Abhinav Madahar · अभिनव ਮਦਾਹਰ
Email: abhinav@abhinavmadahar.com | Phone: +1 (415) 413-8545 | Website: abhinavmadahar.com
Education
Rutgers University – New Brunswick
Bachelor of Science in Computer Science and Mathematics
September 2017 – May 2021
Major Publications
A. Madahar, Natural language edge labelling: Decoupling intent from execution in structured LM reasoning, arXiv preprint, Oct. 6, 2025. arXiv: 2510.04817. [Online]. Available: https://arxiv.org/abs/2510.04817.
A. Madahar, Lateral tree-of-thoughts surpasses ToT by incorporating logically-consistent, low-utility candidates, arXiv preprint, Oct. 1, 2025. arXiv: 2510.01500. [Online]. Available: https://arxiv.org/abs/2510.01500.
Publications
A. Madahar, “Create an international science fund,” Nov. 5, 2025, Submitted as a Nature Comment; Zenodo preprint under CC BY 4.0 license. [Online]. Available: https://doi.org/10.5281/zenodo.17535746.
A. Madahar, “The successor to the modern era: A falsifiable model of global integration,” Nov. 2025, Concept DOI (all versions): 10.5281/zenodo.17503675. [Online]. Available: https://doi.org/10.5281/zenodo.17503725.
A. Madahar, “The recursive collapse of realism in international law,” Oct. 24, 2025, Submitted to Ratio Juris. Preprint available on Zenodo. [Online]. Available: https://doi.org/10.5281/zenodo.17434253.
A. Madahar, “Heavy inference runs: A conceptual unit for high-budget agi reasoning episodes,” AI & Society, Oct. 20, 2025, Open Forum; submission version v1.1.0; under technical check. Preprint submitted to arXiv, pending identifier, submitted.
A. Madahar, “The human time constant: Generational lag and the industrialization of knowledge,” Philosophy & Technology, submitted Oct. 18, 2025; SSRN Working Paper No. 5625310. [Online]. Available: https://ssrn.com/abstract=5625310
A. Madahar, “Freight-first integration: A standard-gauge, ports-anchored path to export-oriented industrialization and political union in East Africa,” World Development (submitted), Oct. 15, 2025. [Online]. Available: https://doi.org/10.5281/zenodo.17395414
A. Madahar, Natural language edge labelling: Decoupling intent from execution in structured LM reasoning, arXiv preprint, Oct. 2025. arXiv: 2510.04817. [Online]. Available: https://arxiv.org/abs/2510.04817
A. Madahar, Lateral tree-of-thoughts surpasses ToT by incorporating logically-consistent, low-utility candidates, arXiv preprint, Oct. 2025. arXiv: 2510.01500. [Online]. Available: https://arxiv.org/abs/2510.01500
A. Madahar, “Inverse electoral accountability: A minimal model of authoritarian retention and its duality with democratic selection,” Comparative Political Studies (under review), Oct. 2025. [Online]. Available: https://doi.org/10.5281/zenodo.17377679
A. Madahar, “Teach AI to think in equilibria: A minimal interdisciplinary core for computer scientists,” AI Magazine (submitted), Oct. 2025. [Online]. Available: https://abhinavmadahar.com/p/artificial-intelligence-pedagogy
A. Madahar, “Reasoning models, self-driving labs, and the path to general cancer control,” Nature Machine Intelligence (submitted), Oct. 2025. Earlier public version: “Toward a general cure for cancer: Why we might need an AGI researcher,” Oct. 8, 2025. [Online]. Available: https://abhinavmadahar.com/p/toward-a-general-cure-for-cancer
A. Madahar, “Poland should skip the AGI arms race — and win at deployment,” The Gradient (Perspective) (submitted), Oct. 2025. Adapted from “How Poland and its Peers Can Advance Their Economies Using Artificial General Intelligence,” Apr. 22, 2025. [Online]. Available: https://abhinavmadahar.com/p/poland-and-agi
A. Madahar, “When AI outruns academia: A model of institutionally perverse incentives for informal learning,” Transition Network Analysis Workshop (TNA@LAK’26) (submitted), Oct. 2025. [Online]. Available: https://doi.org/10.5281/zenodo.17382761
A. Madahar, “Constraining malicious AGI: Exploring the capacities of states and non-state actors to develop and use artificial general intelligence for harm, and the legal barriers that may limit them,” May 25, 2025. [Online]. Available: https://abhinavmadahar.com/p/constraining-malicious-agi
A. Madahar, “If AI is a GPT, are we missing a complement? Examining current AI discourse from a history-informed approach,” Jul. 3, 2025. [Online]. Available: https://abhinavmadahar.com/p/if-ai-is-a-gpt-are-we-missing-a-complement
Research Experience
Industry
Johnson & Johnson — Data Science Co-op
Titusville, New Jersey | Apr. 2020 – Sep. 2020
Mentor: Walter Cedeño, PhD
Oracle Corporation — Data Science Intern
Santa Clara, California | May 2019 – Aug. 2019
Mentor: Dorian Puleri, PhD
Johnson & Johnson — Medical Devices Data Science Intern
Somerville, New Jersey | May 2018 – Aug. 2018
Mentors: Jeff Headd, PhD; Sparkle Russell-Puleri, PhD
Academia
Rutgers University — Department of Computer Science
Research Assistant | May 2020 – Aug. 2020
Principal Investigator: Prof. James Abello Monedero
Research Assistant | Sep. 2018 – May 2019
Principal Investigator: Prof. Sungjin Ahn
Research Assistant | Sep. 2017 – Aug. 2018
Principal Investigator: Prof. Gerard de Melo
Professional Experience
Academia.edu — Software Engineer
San Francisco, California | Jul. 2021 – Mar. 2022
Invited Talks
“Pedagogical Conversational Agents.” AIDeathon (organized by AI Consensus under Minerva University), Aug. 11, 2023.
Service and Outreach
AI Classroom Challenge | Nov. 2023
Judged student proposals on educational use-cases of AI
IEEE Rutgers Student Branch | Sep. 2017 – May 2018
Co-organized weekly undergraduate machine-learning research paper colloquium