PERELMAN: Pipeline for Scientific Literature Meta-Analysis

Sherki, Daniil; Merkulov, Daniil; Savina, Aleksandra; Kikov, Dzhantemir; Parpulov, Dmitry; Ivanov, Alexander; Abakumov, Artem; Muravleva, Ekaterina

doi:10.24963/ijcai.2026/992

PERELMAN: Pipeline for Scientific Literature Meta-Analysis

Daniil Sherki, Daniil Merkulov, Aleksandra Savina, Dzhantemir Kikov, Dmitry Parpulov, Alexander Ivanov, Artem Abakumov, Ekaterina Muravleva

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Demo Track. Pages 8513-8517.

https://doi.org/10.24963/ijcai.2026/992

Keywords

AI: Agent-based and Multi-agent Systems

Abstract

We present PERELMAN (PipEline foR sciEntific Literature Meta-ANalysis), an agentic framework designed to extract specific information from a large corpus of scientific articles to support large-scale literature reviews and meta-analyses. Our central goal is to reliably transform heterogeneous article content into a unified, machine-readable representation. PERELMAN first elicits domain knowledge-including target variables, inclusion criteria, units, and normalization rules-through a structured dialogue with a subject-matter expert. This domain knowledge is then reused across multiple stages of the pipeline and guides coordinated agents in extracting evidence from narrative text, tables, and figures, enabling consistent aggregation across studies. In order to assess reproducibility and validate our implementation, we evaluate the system on the task of reproducing the meta-analysis of layered Li-ion cathode properties LiNi0.8Mn0.1Co0.1O2 (NMC811). We describe our solution, which has the potential to reduce the time required to prepare meta-analyses from months to minutes.

Daniil Sherki, Daniil Merkulov, Aleksandra Savina, Dzhantemir Kikov, Dmitry Parpulov, Alexander Ivanov, Artem Abakumov, and Ekaterina Muravleva. 2026. PERELMAN: Pipeline for Scientific Literature Meta-Analysis. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 8513–8517. https://doi.org/10.24963/ijcai.2026/992

@inproceedings{ijcai2026p992,
  title     = {PERELMAN: Pipeline for Scientific Literature Meta-Analysis},
  author    = {Sherki, Daniil and Merkulov, Daniil and Savina, Aleksandra and Kikov, Dzhantemir and Parpulov, Dmitry and Ivanov, Alexander and Abakumov, Artem and Muravleva, Ekaterina},
  booktitle = {Proceedings of the Thirty-Fifth International Joint Conference on
               Artificial Intelligence, {IJCAI-26}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Diego Calvanese},
  pages     = {8513--8517},
  year      = {2026},
  month     = {8},
  note      = {Demo Track},
  doi       = {10.24963/ijcai.2026/992},
  url       = {https://doi.org/10.24963/ijcai.2026/992},
}

TY  - CPAPER
TI  - PERELMAN: Pipeline for Scientific Literature Meta-Analysis
AU  - Sherki, Daniil
AU  - Merkulov, Daniil
AU  - Savina, Aleksandra
AU  - Kikov, Dzhantemir
AU  - Parpulov, Dmitry
AU  - Ivanov, Alexander
AU  - Abakumov, Artem
AU  - Muravleva, Ekaterina
T2  - Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2026
SP  - 8513
EP  - 8517
DO  - 10.24963/ijcai.2026/992
UR  - https://doi.org/10.24963/ijcai.2026/992
ER  -