Irina Piontkovskaya

Irina Piontkovskaya has 3 papers in our library, in 2023–2026, most of them on Language models & pretraining; often with Eduard Tulchinskii, Kristian Kuznetsov, Laida Kushnareva.

3 papers in our library · 2023–2026

The profile is the library's match of the name; two people with one name are not always told apart.

Affiliations

From the bylines of their papers (OpenAlex, 2026-10-14), ORCID and Wikidata; institutions on papers are matched by machine.

Of its works in our library, with the number of works on each.

The most cited first (OpenAlex and our library).

Intrinsic Dimension Estimation for Robust Detection of AI-Generated Texts (2023 · 79 citations)
Detectors of machine-generated text often fail when the generator, topic domain or writer changes. They also tend to flag essays by non-native speakers as AI-written. The authors look for a property of human writing that stays stable across these conditions. They treat each text sample as a cloud of embedding points…
PanGu-{\Sigma}: Towards Trillion Parameter Language Model with Sparse Heterogeneous Computing (2023 · 40 citations)
PanGu-Σ is a Huawei technical report on training a trillion-parameter language model on limited hardware. Sparse Mixture-of-Experts models usually suffer from unbalanced expert load and costly all-to-all communication, and offloading work to the host CPU tends to reduce throughput. The authors start from PanGu-α…
Unveiling Intrinsic Dimension of Texts: from Academic Abstract to Creative Story (2026)
Intrinsic dimension (ID) of LLM embeddings is often used to study training dynamics, scaling and dataset structure.

6 co-authors with a page, the most papers together first.

Where their works here were published, the most works first, with the editions of those that have a page.

Papers of our library cite their works 72 times (of the 149,439 papers whose bibliographies we read); 20 citing papers with a page here →.

Updated 1 November 2026