activity
20222026
most citedFindings of the The RuATD Shared Task 2022 on Artificial Text Detection in Russian

13 citations · 20 across the 11 of their papers we have counts for

collaborators

13 papers

cs.CL2026

Your Transformer Can Hold Two Thoughts at Once: Evidence of Linear Superposition in LLMs

Pavel Tikhonov, Anton Korznikov, Matvey Mikhalchuk +6

While Large Language Models (LLMs) rely on highly non-linear components, in this work we demonstrate that they exhibit fundamental linearity: when inputs from distinct text streams…

cs.CL2026

What Attention Recalls and Recurrence Controls in Hybrid Language Models

Kirill Afendulev, Alexey Dontsov, Elena Tutubalina +1

Hybrid language models combine attention with a fixed-size recurrent state, but the role of each channel remains unclear. We introduce two cache-level interventions. Split-prefill…

cs.LG2026

Sanity Checks for Sparse Autoencoders: Do SAEs Beat Random Baselines?

Anton Korznikov, Andrey Galichin, Alexey Dontsov +3

Sparse Autoencoders (SAEs) have emerged as a promising tool for interpreting neural networks by decomposing their activations into sparse sets of human-interpretable features. Rece…

cs.CL2026

Anatomy of Unlearning: The Dual Impact of Fact Salience and Model Fine-Tuning

Anna Borisiuk, Andrey Savchenko, Alexander Panchenko +1

Machine Unlearning (MU) enables Large Language Models (LLMs) to remove unsafe or outdated information. However, existing work assumes that all facts are equally forgettable and lar…

cs.LG2025

OrtSAE: Orthogonal Sparse Autoencoders Uncover Atomic Features

Anton Korznikov, Andrey Galichin, Alexey Dontsov +3

Sparse autoencoders (SAEs) are a technique for sparse decomposition of neural network activations into human-interpretable features. However, current SAEs suffer from feature absor…

cs.LG2025

The Rogue Scalpel: Activation Steering Compromises LLM Safety

Anton Korznikov, Andrey Galichin, Alexey Dontsov +3

Activation steering is a promising technique for controlling LLM behavior by adding semantically meaningful vectors directly into a model's hidden states during inference. It is of…