Illia Polosukhin is the founder of NEAR AI, whose company page describes Polosukhin as a co-author of Attention Is All You Need and the creator of the NEAR blockchain. That 2017 paper, which introduced the Transformer, notes that Polosukhin's work on it was performed while at Google Research, and its contribution note says Ashish Vaswani, with Polosukhin, designed and implemented the first Transformer models.12
At a glance
- Current roles
- Founder +1 more
- Companies
- NEAR AI, NEAR Foundation
- Sector
- Company
- Fields
- Language models
- Work listed
- 1
- Sources cited
- 10
- Profile reviewed
Research
Polosukhin's early research papers addressed language understanding over large text collections. Polosukhin co-authored WikiReading (2016), a large-scale natural language understanding task built on Wikipedia, and a 2016 framework for answering questions over long documents. A 2017 paper on TensorFlow Estimators, a high-level machine learning framework, also lists Polosukhin among its authors.456
Polosukhin is a co-author of "Attention Is All You Need" (2017), the paper that introduced the Transformer architecture. The paper's contribution notes say that Ashish Vaswani and Polosukhin designed and implemented the first Transformer models, and that Polosukhin's work was done while at Google Research.2
Program synthesis and NEAR
With Alexander Skidanov, Polosukhin wrote "Neural Program Search" (2018), an algorithm that generates programs from a natural language description and a few input-output examples, and co-authored NAPS, a dataset of human-written problem statements and solutions for program synthesis.78
NEAR AI's company page describes Polosukhin as the creator of the NEAR blockchain and the founder of NEAR AI, which works on verified, user-owned and private AI. Polosukhin and Skidanov's 2025 paper "Proof of Response" presents a mechanism in which a request in a network receives either a response within a known time bound or a proof that a link along the way failed, and it appears among the papers listed by the NEAR ecosystem. The NEAR Foundation lists Polosukhin as a member of its council.19103
Work and contributions
Transformer (Attention Is All You Need) Published research
Co-author of the 2017 paper that introduced the Transformer; the contribution note says Ashish Vaswani, with Polosukhin, designed and implemented the first Transformer models.2
Timeline
- Co-authored WikiReading, a large-scale language understanding task over Wikipedia.4
- Co-authored "Hierarchical Question Answering for Long Documents".5
- Co-authored "Attention Is All You Need", which introduced the Transformer.2
- Co-authored the paper describing TensorFlow Estimators.6
- Posted "Neural Program Search" with Alexander Skidanov as first author.7
- Co-authored NAPS, a natural program synthesis dataset.8
- Posted "Proof of Response" with Alex Skidanov as first author.9
Other documented roles
Roles the sources document that are not confirmed as current.
- Work on the Transformer paper performed while at Google Research, Google Research2
Where they work
- NEAR AI 1 person in the wiki
- NEAR Foundation 1 person in the wiki
Related people
- Arthur ConmyLanguage models
- Ming ZhouLanguage models
- Nanyun (Violet) PengLanguage models
- Luke ZettlemoyerLanguage models
- Myle OttLanguage models
- Liliang RenLanguage models
- Alexander AminiLanguage models
- Dwarkesh PatelLanguage models
- Jyoti AnejaLanguage models
- Sachin MehtaLanguage models
Suggest a correction or removal
Profiles cover public professional work only, and each statement cites a source. If something is out of date, missing or wrong, tell us and link a source. People can also ask for their profile to be corrected or removed.
References
10 sources · profile reviewed Oct 21, 2026 · data updated Oct 21, 2026
- NEAR AI — Company
- Attention Is All You Need (arXiv 1706.03762)
- NEAR Foundation — About (NEAR Foundation Council)
- WikiReading: A Novel Large-scale Language Understanding Task over Wikipedia (arXiv 1608.03542)
- Hierarchical Question Answering for Long Documents (arXiv 1611.01839)
- TensorFlow Estimators: Managing Simplicity vs. Flexibility in High-Level Machine Learning Frameworks (arXiv 1708.02637)
- Neural Program Search: Solving Programming Tasks from Description and Examples (arXiv 1802.04335)
- NAPS: Natural Program Synthesis Dataset (arXiv 1807.03168)
- Proof of Response (arXiv 2502.10637)
- NEAR — Papers