--- title: "Pavel Izmailov — author profile and papers" description: "Pavel Izmailov, Lawrence Livermore National Laboratory: 2 papers on Elamur Research (2018–2023), h-index 19 — affiliations, topics, co-authors and who…" canonical: "https://research.elamur.ai/free/en/authors/pavel-izmailov/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/authors/pavel-izmailov/) [home](https://md.research.elamur.ai/free/en/) / [authors](https://md.research.elamur.ai/free/en/authors/) / Pavel Izmailov # Pavel Izmailov Pavel Izmailov has 2 papers in our library, in 2018–2023; often with [Bowen Baker](https://md.research.elamur.ai/free/en/authors/bowen-baker/), [Collin Burns](https://md.research.elamur.ai/free/en/authors/collin-burns/), [Ilya Sutskever](https://md.research.elamur.ai/free/en/authors/ilya-sutskever/). 2 papers in our library · 2018–2023 [OpenAlex](https://openalex.org/A5069234607) · [ORCID](https://orcid.org/0009-0005-0772-2879) · [**h-index 19**](https://openalex.org/authors/A5069234607) · [43 works](https://openalex.org/works?filter=author.id:A5069234607) · [1,538 citations](https://openalex.org/works?filter=author.id:A5069234607&sort=cited_by_count:desc) OpenAlex, 2026-10-14 [📄 2 works in our library](https://md.research.elamur.ai/free/en/authors/pavel-izmailov/#papers) · [🔗 1,289 citations in our library](https://md.research.elamur.ai/free/en/authors/pavel-izmailov/cited-by/) · [⭐ Paper Star rank #2912](https://md.research.elamur.ai/free/en/paper-stars/page/30/#star-2912) The profile is the library's match of the name; two people with one name are not always told apart. Elsewhere: [Google Scholar](https://scholar.google.com/citations?view_op=search_authors&mauthors=Pavel+Izmailov) · [Semantic Scholar](https://www.semanticscholar.org/search?q=Pavel+Izmailov) · [dblp](https://dblp.org/search?q=Pavel+Izmailov) · [ACL Anthology](https://aclanthology.org/search/?q=Pavel+Izmailov) ## Affiliations [🎓**New York University** New York · United States 2019–2026 · 20 papers*current*](https://md.research.elamur.ai/free/en/universities/new-york-university/) · [🔬**Lawrence Livermore National Laboratory** Livermore · United States 2026 · 1 paper*current*](https://md.research.elamur.ai/free/en/labs/lawrence-livermore-national-laboratory/) · [🎓**Princeton University** Princeton · United States 2026 · 1 paper*current*](https://md.research.elamur.ai/free/en/universities/princeton-university/) · [🎓**University of California, Los Angeles** Los Angeles · United States 2026 · 1 paper*current*](https://md.research.elamur.ai/free/en/universities/university-of-california-los-angeles/) · [🎓**Cornell University** Ithaca · United States 2018–2019 · 5 papers](https://md.research.elamur.ai/free/en/universities/cornell-university/) From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine. 3 more institutions on single bylines are not listed. ## Works by year ### 2020s [**2023** 1 paper](https://md.research.elamur.ai/free/en/papers/burns-2023-weak-strong-generalization-eliciting-strong/) ### 2010s [**2018** 1 paper](https://md.research.elamur.ai/free/en/papers/izmailov-2018-averaging-weights-leads-wider-optima/) ## Papers The most cited first (OpenAlex and our library). 1. **[Averaging Weights Leads to Wider Optima and Better Generalization](https://md.research.elamur.ai/free/en/papers/izmailov-2018-averaging-weights-leads-wider-optima/)** *(2018 · 355 citations)* Deep networks are usually trained with SGD and a decaying learning rate, and the final iterate tends to sit near the edge of a broad region of good solutions rather than in its middle. 2. **[Weak-to-Strong Generalization: Eliciting Strong Capabilities With Weak Supervision](https://md.research.elamur.ai/free/en/papers/burns-2023-weak-strong-generalization-eliciting-strong/)** *(2023 · 163 citations)* Today's alignment methods, such as RLHF, assume humans can judge whether a model's outputs are good. That assumption may fail once models exceed human abilities. The OpenAI authors build a testable stand-in for this situation: a small "weak" model produces labels, and a much larger "strong" pretrained model is… ## Co-authors here 6 co-authors with a page, the most papers together first. - [Bowen Baker](https://md.research.elamur.ai/free/en/authors/bowen-baker/) - [Collin Burns](https://md.research.elamur.ai/free/en/authors/collin-burns/) - [Ilya Sutskever](https://md.research.elamur.ai/free/en/authors/ilya-sutskever/) - [Jan Leike](https://md.research.elamur.ai/free/en/authors/jan-leike/) - [Jeff Wu](https://md.research.elamur.ai/free/en/authors/jeff-wu/) - [Leo Gao](https://md.research.elamur.ai/free/en/authors/leo-gao/) ## Cited by Papers of our library cite their works 1,289 times (of the 140,884 papers whose bibliographies we read); [42 citing papers with a page here →](https://md.research.elamur.ai/free/en/authors/pavel-izmailov/cited-by/).