Papers by Vasily Konovalov

12 papers
Through the Looking Glass: Common Sense Consistency Evaluation of Weird Images (2025.naacl-srw)

Copied to clipboard

Challenge: Existing methods to measure image common sense inconsistentness are difficult to implement because of their complexity.
Approach: They propose a visual commonsense model that leverages large vision-language models to extract atomic facts from images and a compact attention-pooling classifier to fine-tune it over encoded atomic fact.
Outcome: The proposed method outperforms existing methods on the WHOOPS! and WEIRD datasets while maintaining a compact attention-pooling classifier over encoded atomic facts.
Fine-Grained Semantic Comparison of Legal Documents using LLMs (2026.acl-srw)

Copied to clipboard

Challenge: Existing tools for detecting inconsistencies and contradictions in complex regulatory documents rely on character-level diffs.
Approach: They propose a benchmark to evaluate span-aware semantic comparison of legal documents . legDiff is an annotated pair of legal paragraphs that is automatically generated .
Outcome: The proposed benchmark evaluates span-aware semantic comparisons of legal documents . it generates synthetic training data that aligns with the manual annotations and mirrors the structure and label distribution of the benchmark .
DeepPavlov 1.0: Your Gateway to Advanced NLP Models Backed by Transformers and Transfer Learning (2024.emnlp-demo)

Copied to clipboard

Challenge: Open-source framework for using NLP models is released for non-experts . complexity of building, fine-tuning and deploying state-of-the-art models remains a barrier .
Approach: They present DeepPavlov 1.0, an open-source framework for using NLP models . the framework is based on PyTorch and supports HuggingFace transformers .
Outcome: The DeepPavlov 1.0 framework is designed for practitioners with limited knowledge of NLP/ML.
Adaptive Retrieval Without Self-Knowledge? Bringing Uncertainty Back Home (2025.acl-long)

Copied to clipboard

Challenge: Recent adaptive retrieval methods integrate LLMs’ intrinsic knowledge with external information appealing to LLM self-knowledge, but they often neglect efficiency evaluations and comparisons with uncertainty estimation techniques.
Approach: They propose to integrate LLMs’ intrinsic knowledge with external information appealing to LLM self-knowledge but neglect efficiency evaluations and comparisons with uncertainty estimation techniques.
Outcome: The proposed methods outperform complex pipelines in terms of efficiency and self-knowledge while maintaining comparable QA performance.
Will It Still Be True Tomorrow? Multilingual Evergreen Question Classification to Improve Trustworthy QA (2025.emnlp-main)

Copied to clipboard

Challenge: Large Language Models (LLMs) often struggle with question answering due to hallucinated answers.
Approach: They propose a multilingual QA dataset with evergreen labels that can be used to evaluate and train large language models.
Outcome: The proposed model performs well on 12 modern LLMs and EG-E5 classifiers.
LLM-Independent Adaptive RAG: Let the Question Speak for Itself (2025.emnlp-main)

Copied to clipboard

Challenge: Existing methods to retrieve Large Language Models (LLMs) are inefficient and impractical.
Approach: They propose a lightweight adaptive retrieval method that leverages external information to achieve comparable quality while achieving significant efficiency gains.
Outcome: The proposed methods achieve comparable quality while achieving significant efficiency gains on 6 QA datasets.
SPY: Enhancing Privacy with Synthetic PII Detection Dataset (2025.naacl-srw)

Copied to clipboard

Challenge: Historically, Named Entity Recognition (NER) has been employed for PII detection, but PI I entities constitute a subset of NER entities.
Approach: They propose to use Large Language Models to generate a synthetic dataset that emulates real-world PII scenarios and validate its quality.
Outcome: The proposed dataset is validated and provides a benchmark for PII detection.
DeepPavlov Strikes Back: A Toolkit for Improving LLM Reliability and Trustworthiness (2026.eacl-demo)

Copied to clipboard

Challenge: a new version of an open-source library for natural language processing (NLP) is released . the library is easy to use, flexible, and works with many languages.
Approach: They introduce a new version of an open-source library for natural language processing (NLP) it supports both traditional NLP tasks and new tasks needed to enhance LLMs truthfulness and reliability.
Outcome: The new version of DeepPavlov supports both traditional and new tasks . it is easy to use, flexible, and works with many languages .
Efficient Answer Retrieval System (EARS): Combining Local DB Search and Web Search for Generative QA (2024.emnlp-industry)

Copied to clipboard

Challenge: Developing a virtual assistant is crucial for supporting clients as it provides 24/7 assistance . factual questionanswering system is capable of handling all user queries .
Approach: They propose a production-ready factual question answering system that combines local knowledge base search with generative, context-based QA.
Outcome: The proposed system boosts local knowledge base retrieval by 23% . the system is language-agnostic and can be applied to any data domain .
How Much Knowledge Can You Pack into a LoRA Adapter without Harming LLM? (2025.findings-naacl)

Copied to clipboard

Challenge: Low-rank adaptation (LoRA) is a popular training technique for updating or domain-specific adaptation of Large Language Models (LLMs).
Approach: They propose to use low-rank adaptation to incorporate new facts into the LLM without compromising previously learned knowledge.
Outcome: The proposed approach is harmful because the model's performance declines after such fine-tuning.
When Models Lie, We Learn: Multilingual Span-Level Hallucination Detection with PsiloQA (2025.findings-emnlp)

Copied to clipboard

Challenge: Existing hallucination detection benchmarks operate at the sequence level and are limited to English . Existing methods lacking fine-grained, multilingual supervision are limited in English based on the sequence .
Approach: They propose a large-scale, multilingual dataset annotated with span-level hallucinations across 14 languages.
Outcome: The proposed dataset annotated with span-level hallucinations across 14 languages is scalable and cost-efficient.
Multimodal Evaluation of Russian-language Architectures (2026.eacl-long)

Copied to clipboard

Challenge: Multimodal large language models (MLLMs) are at the center of research attention, yet intelligence, limitations, and risks remain insufficiently understood.
Approach: They propose an open multimodal evaluation framework for Russian-spoken architectures . the framework is instruction-based and includes 18 newly constructed evaluation tasks .
Outcome: The proposed framework provides a replicable methodology for constructing multimodal benchmarks in Russian-spoken architectures.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations