2 citations · 4 across the 7 of their papers we have counts for
8 papers
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…
Real-World Transferable Adversarial Attack on Face-Recognition Systems
Andrey Kaznacheev, Matvey Mikhalchuk, Andrey Kuznetsov +2
Adversarial attacks on face recognition (FR) systems pose a significant security threat, yet most are confined to the digital domain or require white-box access. We introduce GaP (…
Inverting Black-Box Face Recognition Systems via Zero-Order Optimization in Eigenface Space
Anton Razzhigaev, Matvey Mikhalchuk, Klim Kireev +3
Reconstructing facial images from black-box recognition models poses a significant privacy threat. While many methods require access to embeddings, we address the more challenging…
LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers
Anton Razzhigaev, Matvey Mikhalchuk, Temurbek Rahmatullaev +4
We introduce methods to quantify how Large Language Models (LLMs) encode and store contextual information, revealing that tokens often seen as minor (e.g., determiners, punctuation…
Universal Adversarial Attack on Aligned Multimodal LLMs
Temurbek Rahmatullaev, Polina Druzhinina, Nikita Kurdiukov +3
We propose a universal adversarial attack on multimodal Large Language Models (LLMs) that leverages a single optimized image to override alignment safeguards across diverse queries…
Your Transformer is Secretly Linear
Anton Razzhigaev, Matvey Mikhalchuk, Elizaveta Goncharova +4
This paper reveals a novel linear characteristic exclusive to transformer decoders, including models such as GPT, LLaMA, OPT, BLOOM and others. We analyze embedding transformations…
OmniFusion Technical Report
Elizaveta Goncharova, Anton Razzhigaev, Matvey Mikhalchuk +6
Last year, multimodal architectures served up a revolution in AI-based approaches and solutions, extending the capabilities of large language models (LLM). We propose an \textit{Om…
The Shape of Learning: Anisotropy and Intrinsic Dimensions in Transformer-Based Models
Anton Razzhigaev, Matvey Mikhalchuk, Elizaveta Goncharova +3
In this study, we present an investigation into the anisotropy dynamics and intrinsic dimension of embeddings in transformer architectures, focusing on the dichotomy between encode…