Papers by Andrey Savchenko
Anatomy of Unlearning: The Dual Impact of Fact Salience and Model Fine-Tuning (2026.findings-acl)
Copied to clipboard
| Challenge: | Existing studies assume that all facts are equally forgettable . popular facts, frequent and widely distributed, may be more deeply embedded than rare ones, making them harder to erase. |
| Approach: | They propose a benchmark to evaluate how unlearning differs between pretrained and supervised fine-tuned models when fact popularity is taken into account. |
| Outcome: | The proposed model is compared with pretrained and SFT models on the forget data and shows that it performs better on both models. |
MADD: Multi-Agent Drug Discovery Orchestra (2025.findings-emnlp)
Copied to clipboard
Gleb Vitalevich Solovev, Alina Borisovna Zhidkovskaya, Anastasia Orlova, Nina Gubina, Anastasia Vepreva, Rodion Golovinskii, Ilya Tonkii, Ivan Dubrovsky, Ivan Gurev, Dmitry Gilemkhanov, Denis Chistiakov, Timur A. Aliev, Ivan Poddiakov, Galina Zubkova, Ekaterina V. Skorb, Vladimir Vinogradov, Alexander Boukhanovsky, Nikolay Nikitin, Andrei Dmitrenko, Anna Kalyuzhnaya, Andrey Savchenko
| Challenge: | Recent advances in artificial intelligence have limited access to wet-lab tools for hit identification . multi-agent systems combine interpretability of LLMs with precision of specialized models and tools . |
| Approach: | They propose a multi-agent system that builds and executes customized hit identification pipelines from natural language queries. |
| Outcome: | The proposed system reduces the complexity of traditional screening methods and improves efficiency. |
WeightLoRA: Keep Only Necessary Adapters (2026.acl-long)
Copied to clipboard
| Challenge: | Low-rank adaptation (LoRA) adds trainable adapters to selected layers, but requires significant memory to train large models and intuition on which layers to add adapters. |
| Approach: | They propose a method which adds trainable adapters to selected layers . they compare weightLoRA with different adaptive approaches to reduce trainable parameters while maintaining consistent or even superior metric values. |
| Outcome: | The proposed method reduces the number of trainable parameters while maintaining the capability to obtain consistent or even superior metric values. |
CRL-Prompt: Contrastive and Reinforcement Learning for Soft Prompt Tuning for Text Classification (2026.acl-srw)
Copied to clipboard
| Challenge: | Manual prompt engineering is time-consuming, non-scalable, and brittle, while current auto-prompting techniques are far from maturity. |
| Approach: | They propose a two-stage method for prompt learning of frozen language models, CRL-Prompt, based on soft prompt initialization followed by contrastive and reinforcement-based refinement. |
| Outcome: | The proposed method achieves consistent improvements over baseline prompt tuning strategies, with gains of up to 2.2% while training fewer than 0.25% of model parameters. |
Hallucination Detection in LLMs with Topological Divergence on Attention Graphs (2026.acl-long)
Copied to clipboard
Alexandra Bazarova, Andrei Volodichev, Aleksandr Yugay, Andrey Shulga, Alina Ermilova, Konstantin Polev, Julia Belikova, Rauf Parchiev, Dmitry Simakov, Maxim Savchenko, Andrey Savchenko, Serguei Barannikov, Alexey Zaytsev
| Challenge: | Large language models (LLMs) are prone to producing so-called hallucinations, i.e., content that is factually or contextually incorrect. |
| Approach: | They propose a TOpology-based HAllucination detector which quantifies the structural properties of graphs induced by attention matrices. |
| Outcome: | The proposed detector achieves state-of-the-art or competitive results on several benchmarks while requiring minimal annotated data and computational resources. |
Leveraging Summarization for Unsupervised Dialogue Topic Segmentation (2024.findings-naacl)
Copied to clipboard
Aleksei Artemiev, Daniil Parinov, Alexey Grishanov, Ivan Borisov, Alexey Vasilev, Daniil Muravetskii, Aleksey Rezvykh, Aleksei Goncharov, Andrey Savchenko
| Challenge: | Existing methods to segment textual data are difficult to handle for noisy spoken dialogues. |
| Approach: | They propose to leverage dialogue summaries for unsupervised topic segmentation . they show that the new approach outperforms state-of-the-art methods in unsupervised segmentation and requires less setup . |
| Outcome: | The proposed approach outperforms state-of-the-art methods in unsupervised topic segmentation and requires less setup. |
The Silence of the Facts: Popularity as a Barrier to Machine Unlearning (2026.acl-srw)
Copied to clipboard
| Challenge: | Existing unlearning methods assume that all facts are equally challenging to forget . large models struggle more to forget popular entities, damaging related knowledge in the process . |
| Approach: | They build a benchmark to investigate whether fact popularity influences the efficiency of LLM unlearning. |
| Outcome: | The proposed benchmark compares state-of-the-art models on a set of models of different sizes. |
Lost in Translation: Chemical Language Models and the Misunderstanding of Molecule Structures (2024.findings-emnlp)
Copied to clipboard
Veronika Ganeeva, Andrey Sakhovskiy, Kuzma Khrabrov, Andrey Savchenko, Artur Kadurin, Elena Tutubalina
| Challenge: | chemistry and natural language processing (NLP) have advanced drug discovery. |
| Approach: | They propose a framework for assessment of Chemistry LMs of different natures that relies on augmentations that preserve an underlying chemical. |
| Outcome: | The proposed framework relies on augmentations that preserve an underlying chemical, such as kekulization and cycle replacements. |
Ad Lingua: Text Classification Improves Symbolism Prediction in Image Advertisements (2020.coling-main)
Copied to clipboard
Andrey Savchenko, Anton Alekseev, Sejeong Kwon, Elena Tutubalina, Evgeny Myasnikov, Sergey Nikolenko
| Challenge: | a recent study shows that image-based symbols are insufficient for symbolism prediction in visual advertising . a new method is proposed to help understand image advertisements . |
| Approach: | They propose a multimodal image-based classifier and object detection classifier for symbols . they propose 'symbolic' annotation tasks to help users understand ads' |
| Outcome: | The proposed system establishes state-of-the-art in symbolism prediction. |
3MDBench: Medical Multimodal Multi-agent Dialogue Benchmark (2025.emnlp-main)
Copied to clipboard
Ivan Sviridov, Amina Miftakhova, Tereshchenko Artemiy Vladimirovich, Galina Zubkova, Pavel Blinov, Andrey Savchenko
| Challenge: | Large Vision-Language Models (LVLMs) are being explored in medicine but their ability to conduct complex real-world telemedicine consultations remains underexplored. |
| Approach: | They propose to use large vision-language models to conduct telemedicine consultations using a framework that simulates patient variability and evaluates diagnostic accuracy and dialogue quality via Assessor Agent. |
| Outcome: | The proposed framework compares diagnostic strategies for open and closed-source LVLMs and shows that multimodal dialogue improves F1 score by 6.5% over non-dialogue settings. |
ATGen: A Framework for Active Text Generation (2025.acl-demo)
Copied to clipboard
Akim Tsvigun, Daniil Vasilev, Ivan Tsvigun, Ivan Lysenko, Talgat Bektleuov, Aleksandr Medvedev, Uliana Vinogradova, Nikita Severin, Mikhail Mozikov, Andrey Savchenko, Ilya Makarov, Grigorev Rostislav, Ramil Kuleev, Fedor Zhdanov, Artem Shelmanov
| Challenge: | Despite the surging popularity of natural language generation tasks, the application of active learning (AL) to NLG has been limited. |
| Approach: | They propose a framework that bridges AL with text generation tasks and provides a unified platform for smooth implementation and benchmarking of novel AL strategies tailored to NLG tasks. |
| Outcome: | The proposed framework simplifies AL-empowered annotation in NLG tasks using both human annotators and automatic annotation agents based on large language models (LLMs). |
LATTE: Learning Aligned Transactions and Textual Embeddings for Bank Clients (2025.emnlp-industry)
Copied to clipboard
Egor Fadeev, Dzhambulat Mollaev, Aleksei Shestov, Dima Korolev, Omar Zoloev, Ivan A Kireev, Andrey Savchenko, Maksim Makarenko
| Challenge: | Large language models (LLMs) are computationally expensive and impractical for real-world pipelines. |
| Approach: | They propose a contrastive learning framework that aligns raw event embeddings with description-based semantic embedds from frozen LLMs. |
| Outcome: | The proposed framework outperforms state-of-the-art techniques for learning event sequence representations on real-world financial datasets while remaining deployable in latency-sensitive environments. |