1 citations · 2 across the 11 of their papers we have counts for
11 papers
Closing Cost-Quality Gap in Document VLMs: Difficulty-Aware Data Curation and Quality-Adjusted Deployment Economics
Maksim Evdokimov, Matvey Ivanov, Dmitrii Tsiupin +3
Extracting structured fields from hundreds of millions of documents annually remains costly in regulated industries: bespoke OCR cascades cover only a fraction of workflows, privac…
From Production Traffic to Post-Training: Building a Self-Hosted LLM That Covers the Corporate Request Mix
Olga Tsymboi, Dmitrii Stoianov, Ramil Latypov +11
Data-residency constraints force enterprises to self-host LLMs, but continuous adoption of newer models without decommissioning their predecessors expands the serving fleet, fragme…
Learning Selective LLM Autonomy from Copilot Feedback in Enterprise Customer Support Workflows
Nikita Borovkov, Elisei Rykov, Olga Tsymboi +4
We present a deployed system that automates end-to-end customer support workflows inside an enterprise Business Process Management (BPM) platform. The approach is scalable in produ…
T-pro 2.0: An Efficient Russian Hybrid-Reasoning Model and Playground
Dmitrii Stoianov, Danil Taranets, Olga Tsymboi +12
We introduce T-pro 2.0, an open-weight Russian LLM for hybrid reasoning and efficient inference. The model supports direct answering and reasoning-trace generation, using a Cyrilli…
Kaleidoscope: In-language Exams for Massively Multilingual Vision Evaluation
Israfel Salazar, Manuel Fernández Burda, Shayekh Bin Islam +42
The evaluation of vision-language models (VLMs) has mainly relied on English-language benchmarks, leaving significant gaps in both multilingual and multicultural coverage. While mu…
Towards Foundation Time Series Model: To Synthesize Or Not To Synthesize?
Kseniia Kuvshinova, Olga Tsymboi, Alina Kostromina +2
The industry is rich in cases when we are required to make forecasting for large amounts of time series at once. However, we might be in a situation where we can not afford to trai…
Sparse and Transferable Universal Singular Vectors Attack
Kseniia Kuvshinova, Olga Tsymboi, Ivan Oseledets
The research in the field of adversarial attacks and models' vulnerability is one of the fundamental directions in modern machine learning. Recent studies reveal the vulnerability…
General Lipschitz: Certified Robustness Against Resolvable Semantic Transformations via Transformation-Dependent Randomized Smoothing
Dmitrii Korzh, Mikhail Pautov, Olga Tsymboi +1
Randomized smoothing is the state-of-the-art approach to construct image classifiers that are provably robust against additive adversarial perturbations of bounded magnitude. Howev…
Translate your gibberish: black-box adversarial attack on machine translation systems
Andrei Chertkov, Olga Tsymboi, Mikhail Pautov +1
Neural networks are deployed widely in natural language processing tasks on the industrial scale, and perhaps the most often they are used as compounds of automatic machine transla…
NAG-GS: Semi-Implicit, Accelerated and Robust Stochastic Optimizer
Valentin Leplat, Daniil Merkulov, Aleksandr Katrutsa +3
Classical machine learning models such as deep neural networks are usually trained by using Stochastic Gradient Descent-based (SGD) algorithms. The classical SGD can be interpreted…
Denoising Score Matching with Random Fourier Features
Tsimboy Olga, Yermek Kapushev, Evgeny Burnaev +1
The density estimation is one of the core problems in statistics. Despite this, existing techniques like maximum likelihood estimation are computationally inefficient due to the in…