Category: Университетская жизнь /

Translation. Region: Russian Federation –

Source: Novosibirsk State University –

An important disclaimer is at the bottom of this article.

From March 7 to 12, the final round of the 22nd All-Russian School Chemistry Tournament was held for the eighth time at Novosibirsk State University (NSU). This year, 136 teams from 28 regions of the country—from Krasnodar Krai and Belgorod Oblast to Siberia—participated in the tournament's regional qualifying rounds. The top 20 teams from 11 regions of Russia were invited to the final round, competing for the title of winner.

The tournament format consists of a series of battles pitting three or four teams against each other. One team presents a problem the students have been working on throughout the school year. The second and third teams assume the roles of opponent and reviewer, evaluating the completeness of the solution, its feasibility, and the presenter's ability to convey their idea to the audience.

The tournament featured 29 scientific battles, which determined the top three teams and recognized outstanding players. The best first-degree player was Anastasia Abramova from the Small Academy, Krasnodar. Second-degree diplomas were awarded to Gleb Raspopov (Secondary School No. 471, Vyborg District, St. Petersburg), Dmitry Prikhodko and Nikolai Kravchenko (NSU Specialized Educational Center), and Ruzana Ryazanova (LingTech No. 28, Kirov). Third-degree diplomas were awarded to Egor Anufriev (Southern Federal District Specialized Educational Center), Mikhail Pupkov and Maria Chernosova (Gornostay Educational Center), Ilya Burashnikov (ChOMLI, Chelyabinsk), and Eva Daus (Small Academy).

Diplomas were also awarded to the best speakers of the tournament: Alexandra Shchitova (Lyceum No. 130, Novosibirsk), Semyon Konstantinov (NSU SUNC), Denis Revko (MAOU "Gymnasium No. 1"), opponents: Alina Kovtun (KOGAOU DO TsDOOSH, Kirov), Ulyana Kuznetsova (MBOU OTs "Sodruzhestvo", Voronezh), Alexandra Smirnova (MOAU LInTech No. 28, Kirov), Vitaly Sunegin (SPb GBOU "Lyceum No. 280 named after M.Yu. Lermontov", St. Petersburg), and reviewers: Semyon Konstantinov (NSU SUNC), Victoria Bobynina (Gymnasium No. 5, Belgorod), Sofia Korkina (MAOU "Gymnasium No. 1"), Evelina Davydenko (MBOU OTs "Sodruzhestvo", Voronezh), Vasily Zinoviev (St. Petersburg State Budgetary Educational Institution "Lyceum No. 280 named after M.Yu. Lermontov", St. Petersburg), Ekaterina Podchasova (Specialized Educational and Scientific Center "Academic Gymnasium named after D.K. Faddeev", St. Petersburg), Timofey Tsymbal (MBOU DO "Small Academy", Krasnodar).

All awarded teams and participants received prizes, and the winners and runners-up in the individual competition were invited to the NSU Specialized Educational and Scientific Center Summer School; 17 invitations were issued in total. Following the tournament's established tradition, the jury awarded 47 students special nominations (it's customary at the tournament to recognize funny moments and award participants with commemorative certificates).

The tournament program also included tours of NSU institutes, such as the Novosibirsk Institute of Organic Chemistry (NIOC) SB RAS and the Institute of Chemical Biology and Fundamental Medicine SB RAS, as well as a visit to the "Evolution of the Earth" research and education center. Tours of research institutes have long been popular among participants, but this time, the NIOC tour nearly doubled its scheduled time due to the abundance of questions from participants.

This year, the organizing committee is especially pleased to recognize the successes of teams who are returning to the tournament and who, year after year, demonstrate an increasingly high level of preparation and development not only in chemistry but also in other competencies, such as the ability to present their solutions and to accept and process criticism.

The four teams that advanced to the tournament finals presented their best solutions. Team "Carborundum," a team from Lyceum No. 130 and the Specialized Scientific Center of Novosibirsk State University, presented a solution to a problem involving dyeing fur and leather with various dyes obtained in Paleolithic conditions. Third place went to the team "Ugarnye Rebyata" from LenTech No. 28 in Kirov, demonstrating the influence of solvent selection on the reaction mechanism. Team "Pyrotechnics," from Gymnasium No. 1 in Novosibirsk, took second place in the team rankings, presenting a problem that involved describing the properties of chemical glassware that influence the reaction.

The jury for the final chemical battle included two doctors and six candidates of science, experts in physical, analytical, inorganic, and biochemical fields, as well as an expert in kinetics and catalysis. The tournament jury traditionally consisted primarily of postgraduate students from the Faculty of Natural Sciences (FNS) of NSU and candidates of science—current researchers at the Siberian Branch of the Russian Academy of Sciences (SB RAS). This year's jury also included two deputy directors of SB RAS institutes—the Institute of Chemical Biology and Fundamental Medicine (ICBFM SB RAS) and the Institute of Inorganic Chemistry (IIC SB RAS). In turn, tournament alumni regularly join the pool of enthusiastic tournament participants at NSU, not only as volunteers—ten former winners and runners-up are among the tournament's leading bouts.

"If we talk about the main features of this year's tournament, we can see the strong preparation of teams from various cities—the so-called 'dominance' of teams from Novosibirsk is definitely no longer evident. Two of the three prize winners are not from Novosibirsk. This demonstrates that the tournament is truly all-Russian. The level of the participants is also traditionally high. In recent years, I've seen the emergence of a large number of strong and ambitious individuals who are capable of saving their teams with their individual efforts, achieving high scores and making a lasting impression on the audience and the jury," noted Mikhail Lyulyukin, PhD, Senior Researcher at the Institute of Catalysis SB RAS, and Senior Lecturer at the NSU Natural Sciences Department.

The winner of the tournament was the "Khiminol" team (Malaya Akademiia, Krasnodar), who chose to present their "Output Problem" on the influence of various factors on current efficiency during electrolysis. Timofey Tsymbal, captain of the winning team, shared his impressions of the tournament:

"I remember presenting my papers the most. It's a great joy to tackle a problem knowing it's been solved well and you've thoroughly understood the topic. However, there's another side to the coin. Sometimes you have to present a problem that another team member has already solved. And in that case, you have to thoroughly understand their solution in literally 20 minutes. At the 22nd Chemistry Tournament, I experienced both scenarios firsthand, and both left a lasting impression on me.

Timofey explained that this was the students' third time attending the tournament; the previous two years of participation had yielded no success. This time, they were tempted to back out, but they persevered and made the right decision.

"Since our team has been participating in the tournament for three years, we've gained enough experience to navigate the tournament schedule well. This was probably one of the decisive factors in our victory. Here, I'd like to add some advice to young teams: don't give up if something doesn't work out. A winner is someone who has tried many times," Timofey emphasized.

According to Timofey, the chemistry tournament is a unique event, radically different from Olympiads. It evaluates not only the participant's theoretical knowledge, but also their analytical thinking, ability to quickly and accurately answer questions, and the practical skills applied in solving the problem.

"Tournaments like these provide a wonderful opportunity for schoolchildren to go beyond the school curriculum, practice research, and learn constructive communication with sound reasoning and logical connections. A side effect can also be an expansion of their social circle by meeting like-minded people from different cities across the country," concluded Mikhail Lyulyukin.

The tournament was held with the support of the Institute of Chemical Biology and Fundamental Mechanics SB RAS, the Institute of Chemical Solid State Mechanics SB RAS, the Institute of Catalysis SB RAS, the Institute of Organic Chemistry SB RAS, the Institute of Inorganic Chemistry SB RAS, and TD GraSS LLC. The tournament organizing committee expresses its gratitude to Novosibirsk State University, and in particular the Faculty of Natural Sciences and the Specialized Educational and Scientific Center of NSU, for their assistance in hosting the event.

Please note: This information is raw content obtained directly from the source. It represents an accurate account of the source's assertions and does not necessarily reflect the position of MIL-OSI or its clients.

Translation. Region: Russian Federation –

Source: Novosibirsk State University –

An important disclaimer is at the bottom of this article.

A team of researchers from Novosibirsk State University won first place in the international scientific competition SemEval-2026 Task 8 "MTRAGEval: Evaluating Multi-Turn RAG Conversations." The team was led by NSU Associate Professor of Industrial Engineering and a research fellow at the Laboratory of Applied Digital Technologies. Faculty of Mechanics and Mathematics of NSUAssociate Professor Ivan Bondarenko. The results of the competition will be presented at the world's largest conference on computational linguistics, ACL, which will be held in the summer of 2026.

The competition was hosted by IBM and consisted of three tracks. The NSU team participated in Task B—a task of generating answers to user questions based on provided reference documents and the history of a multi-step dialogue. Of the 26 participating teams, the NSU team took first place, achieving a quality metric of 0.7827 (conditioned harmonic mean), significantly exceeding the organizers' best baseline result (0.6390) by 14.4 percentage points.

SemEval (Semantic Evaluation) is an annual international workshop on methods and algorithms for computational semantics, held for over 20 years. This event hosts competitions in various areas of computational linguistics. This year, SemEval presented 13 challenging research problems to participants. One of the most interesting and significant problems was Task 8, which assessed the performance of RAG (Retrieval-Augmented Generation) systems in multi-step dialogues. Such RAG systems address a key issue with modern large language models: their limited worldview and the difficulty of adapting them to specialized subject areas. The "knowledge" of a large language model is limited to the training set and does not include relevant or domain-specific information. RAG integrates language models with external knowledge bases, enabling them to find and use relevant information when generating responses.

"Our team proposed three key approaches that secured our victory in the competition. The first involved iteratively improving the system prompt using an LLM agent. We developed a multi-agent system in which a large Gemini neural network analyzes the model's performance and suggests improvements to the system prompt. This process is repeated iteratively until a plateau is reached. The second approach involved using in-context learning, in which the model learns to perform a task based on several examples of correct solutions to the problem provided in the input context. For each problem category, the researchers selected the most typical examples using the medoid method in a metric embedding space. These examples were added to the prompt to demonstrate the correct behavior of the model. This approach consistently demonstrated the best results," explained Ivan Bondarenko.

The researchers created several network algorithms using both approaches and evaluated their advantages before deciding to combine them. Among a variety of ensemble methods, they chose a method using a judge neural network that would select the best ensemble response in each case. The team combined seven disparate language models (Gemini-3-Pro-Preview, GLM-4.6, Llama-3.3-70B-Instruct, Qwen3-235B-A22B-Instruct, Claude 4.5 Haiku, Qwen2.5-32B-Instruct, and their own model, Meno-Lite-0.1) and used GPT-4o-mini to select the best response in each case. The diversity of models and approaches provided an additional boost in quality.

"The Meno-Lite-0.1 model, our team's own development based on Qwen2.5-7B-Instruct, deserves special attention. This compact model with 7 billion parameters was specifically retrained for use in RAG pipelines on a corpus of Russian- and English-language educational data. Despite its small size, Meno-Lite demonstrated performance comparable to significantly larger models and contributed to the ensemble's final result," explained Ivan Bondarenko.

The NSU team that participated in the competition included current and former NSU students and staff: Mikhail Kulakov, a master's student in the machine learning program implemented jointly with the School of Data Analysis and the Faculty of Mathematics and Mechanics of NSU; Ivan Chernov, a fourth-year student at the NSU Institute of Intelligent Robotics; Mikhail Komarov, a graduate of the NSU Institute of Intelligent Robotics and chief engineer of the RAGU open source project; Oleg Sedukhin, a graduate of the NSU Faculty of Information Technology; and Roman Derunets, a graduate of the NSU Institute of Intelligent Robotics and a participant in the Meno project.

A scientific paper describing their proposed solution has been submitted for peer review and will be presented at the ACL (Association for Computational Linguistics) conference, the world's largest scientific forum on computational linguistics. Ivan Bondarenko emphasized that the results obtained are already being used in the development of the university's internal project, Meno, an intelligent system based on RAG technologies. The methods developed by the team members can be used to improve the quality of dialog systems that work with external knowledge bases, including corporate and educational applications.

Material prepared by: Elena Panfilo, NSU press service

Please note: This information is raw content obtained directly from the source. It represents an accurate account of the source's assertions and does not necessarily reflect the position of MIL-OSI or its clients.