Picture for Ivan Oseledets

Ivan Oseledets

AIRI, Skolkovo Institute of Science and Technology

Test-Time Reasoning Through Visual Human Preferences with VLMs and Soft Rewards

Add code
Mar 25, 2025
Viaarxiv icon

I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders

Add code
Mar 24, 2025
Viaarxiv icon

Exploring specialization and sensitivity of convolutional neural networks in the context of simultaneous image augmentations

Add code
Mar 05, 2025
Viaarxiv icon

Combining Flow Matching and Transformers for Efficient Solution of Bayesian Inverse Problems

Add code
Mar 03, 2025
Viaarxiv icon

M3DA: Benchmark for Unsupervised Domain Adaptation in 3D Medical Image Segmentation

Add code
Feb 24, 2025
Figure 1 for M3DA: Benchmark for Unsupervised Domain Adaptation in 3D Medical Image Segmentation
Figure 2 for M3DA: Benchmark for Unsupervised Domain Adaptation in 3D Medical Image Segmentation
Figure 3 for M3DA: Benchmark for Unsupervised Domain Adaptation in 3D Medical Image Segmentation
Figure 4 for M3DA: Benchmark for Unsupervised Domain Adaptation in 3D Medical Image Segmentation
Viaarxiv icon

LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers

Add code
Feb 20, 2025
Viaarxiv icon

FLAME: Flexible LLM-Assisted Moderation Engine

Add code
Feb 13, 2025
Viaarxiv icon

Screener: Self-supervised Pathology Segmentation Model for 3D Medical Images

Add code
Feb 12, 2025
Figure 1 for Screener: Self-supervised Pathology Segmentation Model for 3D Medical Images
Figure 2 for Screener: Self-supervised Pathology Segmentation Model for 3D Medical Images
Figure 3 for Screener: Self-supervised Pathology Segmentation Model for 3D Medical Images
Figure 4 for Screener: Self-supervised Pathology Segmentation Model for 3D Medical Images
Viaarxiv icon

Spread them Apart: Towards Robust Watermarking of Generated Content

Add code
Feb 11, 2025
Viaarxiv icon

MaxInfo: A Training-Free Key-Frame Selection Method Using Maximum Volume for Enhanced Video Understanding

Add code
Feb 05, 2025
Viaarxiv icon