影响力指数

13.4/100

超过 41.2%

全站排名 #37,843

发表论文3 篇

平均评分4.9

年均产出1.5 篇/年

Viacheslav Surkov

MS student@EPFL - EPF Lausanne·瑞士·OpenReview

研究方向

Deep Learning · Mechanistic Interpretability · Safe ML

Steering Diffusion Transformers with Sparse Autoencoders

ICLR 2026Withdrawn

One-Step is Enough: Sparse Autoencoders for Text-to-Image Diffusion Models

NeurIPS 2025Poster

Unpacking SDXL Turbo: Interpreting Text-to-Image Models with Sparse Autoencoders

ICLR 2025Withdrawn

合作者 (7)

Caglar Gulcehre

Justin Deschenaux

Mikhail Terekhov