Polina Druzhinina

4 papers · 2024–2026 · 4 venues · across top CS/AI venues

Authors are matched by name, so researchers who share a name ("Yang Liu", "Wei Wang") appear as a single profile and their paper counts, venues and achievements are combined. How identity is handled.

Achievements

Jump to papers ↓

🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🚀 Conference Pioneer

Venues

AAAI (1) ACL (1) MICCAI (1) NAACL (1)

Top co-authors

Anton Razzhigaev (3) Ivan Oseledets (2) Laida Kushnareva (1) Elena Tutubalina (1) Andrey V. Galichin (1) Elizaveta Goncharova (1) Maxim Sharaev (1) Kristian Kuznetsov (1) Maria Boyko (1) Andrey Kuznetsov (1)

Research topics

Artificial Intelligence (3) Machine Learning (3) Core AI (3) Core Methods (3) Deep Learning (1) Natural Language Processing (1) Architectures (1) Resources & Methods (1)

Keywords

sparse autoencoder (2) large language model (2) feature extraction (1) contextual information (1) mechanistic interpretability (1) residual stream (1) writing style (1) token-level analysis (1) context memory (1) feature steering (1) logit len (1) punctuation token (1) artificial text detection (1) transformer architecture (1) reasoning feature (1) representation learning (1)

Papers

I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders AAAI 2026 Feature-Level Insights into Artificial Text Detection with Sparse Autoencoders ACL 2025 LLM-Microscope: Uncovering the Hidden Role of Punctuation in Context Memory of Transformers NAACL 2025 MEGFormer: enhancing speech decoding from brain activity through extended semantic representations MICCAI 2024