🧑‍🔬 Max Tegmark

Affiliations

• Massachusetts Institute of Technology · Cambridge, United States — current · 2004–2026 on 228 papers · employment (Wikidata)
• The NSF AI Institute for Artificial Intelligence and Fundamental Interactions · Cambridge, United States — current · 2021–2026 on 25 papers
• Theiss Research · La Jolla, United States — 2016–2020 on 3 papers
• Foundational Questions Institute · Decatur, United States — 2007–2014 on 2 papers
• University of California, Berkeley · Berkeley, United States — 1993–2013 on 9 papers · education (Wikidata)
• University of Pennsylvania · Philadelphia, United States — 1997–2006 on 98 papers
• Institute for Advanced Study · Princeton, United States — 1996–2001 on 48 papers
• Max Planck Institute for Physics · Munich, Germany — 1994–1998 on 21 papers
• Max Planck Institute for Astrophysics · Garching, Germany — 1995–1998 on 11 papers
• Max Planck Society · Munich, Germany — 1995–1997 on 4 papers

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 14 more institutions on single bylines are not listed.

Topics

Language models & pretraining1Efficient models & compression1

Works by year

2020s

Papers

The most cited first (OpenAlex and our library).

1. The Geometry of Truth: Emergent Linear Structure in Large Language Model Representations of True/False Datasets (2023 · 160 citations)
   LLMs sometimes state things that are false even when they seem to "know" better. Earlier attempts to detect truthfulness by probing internal activations generalized poorly, which raised doubts about whether those probes found truth or only features that correlate with it. Marks and Tegmark study this question in…
2. The Quantization Model of Neural Scaling (2023 · 27 citations)
   The paper tackles a puzzle about scaling. Average loss falls smoothly as a power law when models and datasets grow, yet specific abilities often seem to switch on abruptly. To reconcile the two, the authors propose the Quantization Hypothesis. It holds that prediction tasks rely on a countable set of discrete skills…

Also in our library

8 more works of theirs are in our library without a page here yet; they will appear as we publish them.

Cited by

Papers of our library cite their works 270 times (of the 140 884 papers whose bibliographies we read); 13 citing papers with a page here →.

This page in Markdown → https://md.research.elamur.ai/free/en/authors/max-tegmark/