Demonstration venue · read-only. Every page can be browsed; the buttons that would change it are switched off. Create an account to run TaxoReview on your own data. A simulation from public publication records: the people listed were not invited and did not agree to review.
VLDB 2026 Research / reviewers in the wild / expert

Daniil Cherniavskii

dblp:301/9022 · DBLP profile ↗
← Back
4ranked-venue papers
0as first author
4since 2021 · last 2023
—ORCID · none

Domains — the database's venue-derived domains; a paper can count in several

Artificial intelligence and machine learning · 4 · 4 since 2021Graphics, computer vision, multimedia, augmented reality and games · 1 · 1 since 2021

Expertise — from the expertise taxonomy: the topics of the expert's papers under the CCF categories. A weight counts papers with recency: 1 for a paper about the topic, 0.3 when the topic is its context, halved every five years.

Artificial intelligence
3 papers
Representation and self-supervised learning · 70% Information extraction and text analysis · 18% Language models and text generation · 12%
Network and information security
1 paper
Digital forensics and information hiding · 100%

Topics — the 6 heaviest of 8, each with the papers that count most for it

TopicWeightPapersLastEvidence papers
Machine learning › Representation and self-supervised learning › representation learning
dimensionality reduction		 0.712023
Learning topology-preserving data representations · ICLR 2023
Machine learning › Representation and self-supervised learning
topology preservation		 0.712023
Learning topology-preserving data representations · ICLR 2023
Machine learning › Representation and self-supervised learning › representation learning › topological representation learning
topology-preserving representation		 0.712023
Learning topology-preserving data representations · ICLR 2023
Digital forensics and information hiding › synthetic media detection
machine-generated text detection		 0.712023
Intrinsic Dimension Estimation for Robust Detection of AI-Generated Texts · NeurIPS 2023
Natural language and speech › Language models and text generation
AI-generated text		 0.212023
Intrinsic Dimension Estimation for Robust Detection of AI-Generated Texts · NeurIPS 2023
Natural language and speech › Language models and text generation
large language model		 0.112021
Artificial Text Detection via Examining the Topology of Attention Maps · EMNLP (1) 2021

Methods — techniques the papers use, named apart from their topics

manifold learning · 1.3intrinsic dimension estimation · 1.3topological data analysis · 1.2persistent homology · 0.7attention map analysis · 0.5
YearPublicationVenuePosition
2023 Learning topology-preserving data representations
Ilya Trofimov, Daniil Cherniavskii, Eduard Tulchinskii, Nikita Balabin, Evgeny Burnaev, Serguei Barannikov
 ICLR2
2023 Topological Data Analysis for Speech Processing
abstract
International audience
Eduard Tulchinskii, Kristian Kuznetsov, Laida Kushnareva, Daniil Cherniavskii, Serguei Barannikov, Irina Piontkovskaya, Sergey I. Nikolenko, Evgeny Burnaev
 INTERSPEECH4
2023 Intrinsic Dimension Estimation for Robust Detection of AI-Generated Texts
abstract
Rapidly increasing quality of AI-generated content makes it difficult to distinguish between human and AI-generated texts, which may lead to undesirable consequences for society. Therefore, it becomes increasingly important to study the properties of human texts that are invariant over text domains and various proficiency of human writers, can be easily calculated for any language, and can robustly separate natural and AI-generated texts regardless of the generation model and sampling method. In this work, we propose such an invariant of human texts, namely the intrinsic dimensionality of the manifold underlying the set of embeddings of a given text sample. We show that the average intrinsic dimensionality of fluent texts in natural language is hovering around the value $9$ for several alphabet-based languages and around $7$ for Chinese, while the average intrinsic dimensionality of AI-generated texts for each language is $\approx 1.5$ lower, with a clear statistical separation between human-generated and AI-generated distributions. This property allows us to build a score-based artificial text detector. The proposed detector's accuracy is stable over text domains, generator models, and human writer proficiency levels, outperforming SOTA detectors in model-agnostic and cross-domain scenarios by a significant margin.
Eduard Tulchinskii, Kristian Kuznetsov, Laida Kushnareva, Daniil Cherniavskii, Sergey I. Nikolenko, Evgeny Burnaev, Serguei Barannikov, Irina Piontkovskaya
 NeurIPS4
2021 Artificial Text Detection via Examining the Topology of Attention Maps
abstract
Laida Kushnareva, Daniil Cherniavskii, Vladislav Mikhailov, Ekaterina Artemova, Serguei Barannikov, Alexander Bernstein, Irina Piontkovskaya, Dmitri Piontkovski, Evgeny Burnaev. Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 2021.
Laida Kushnareva, Daniil Cherniavskii, Vladislav Mikhailov, Ekaterina Artemova, Serguei Barannikov, Alexander V. Bernstein, Irina Piontkovskaya, Dmitri Piontkovski, Evgeny Burnaev
 EMNLP (1)2