collaborators

10 papers

cs.CL2026

Progressive Cramming: Reliable Token Compression and What It Reveals

Dmitrii Tarasov, Timofei Lashukov, Elizaveta Goncharova +1

Token cramming compresses sequences into learned embeddings with near-perfect reconstruction, but fixed token budgets and 99\% accuracy thresholds leave it unclear whether residual…

cs.CL2026

SONAR-LLM: Autoregressive Transformer that Thinks in Sentence Embeddings and Speaks in Tokens

Nikita Dragunov, Temurbek Rahmatullaev, Elizaveta Goncharova +5

The recently proposed Large Concept Model (LCM) generates text by predicting a sequence of sentence-level embeddings and training with either mean-squared error or diffusion object…

cs.AI2026

NoReGeo: Non-Reasoning Geometry Benchmark

Irina Abdullaeva, Anton Vasiliuk, Elizaveta Goncharova +4

We present NoReGeo, a novel benchmark designed to evaluate the intrinsic geometric understanding of large language models (LLMs) without relying on reasoning or algebraic computati…

cs.LG2025

Simple Vision-Language Math Reasoning via Rendered Text

Matvey Skripkin, Elizaveta Goncharova, Andrey Kuznetsov

We present a lightweight yet effective pipeline for training vision-language models to solve math problems by rendering LaTeX encoded equations into images and pairing them with st…

cs.CL2025

Sentence-Anchored Gist Compression for Long-Context LLMs

Dmitrii Tarasov, Elizaveta Goncharova, Kuznetsov Andrey

This work investigates context compression for Large Language Models (LLMs) using learned compression tokens to reduce the memory and computational demands of processing long seque…

cs.CV2025

Image Reconstruction as a Tool for Feature Analysis

Eduard Allakhverdov, Dmitrii Tarasov, Elizaveta Goncharova +1

Vision encoders are increasingly used in modern applications, from vision-only models to multimodal systems such as vision-language models. Despite their remarkable success, it rem…