Veniamin Veselovsky
5 papers · 2024–2025 · 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 ↓+1 more ↓ Show less ↑
🌍 Conference Polyglot · Bronze 🌉 Interdisciplinary Bridge · Bronze 🗺️ Taxonomy Completionist (15) 🐣 Hot Topic Early Bird ❓ The Questioner
📈
Trend Setter
Top co-authors
Research topics
Artificial Intelligence
(5)
Core AI
(5)
Deep Learning
(2)
Machine Learning
(2)
Natural Language Processing
(2)
Resources & Methods
(2)
Security & Privacy
(1)
Architectures
(1)
Core Methods
(1)
Generation
(1)
Learning Types · Deep Learning
(1)
Learning Types · Machine Learning
(1)
Optimization & Theory
(1)
Privacy
(1)
Keywords
multilingual language model
(2)
transformer model
(2)
benchmark evaluation
(1)
embedding space
(1)
cross-lingual transfer
(1)
model evaluation
(1)
instruction following
(1)
semantic representation
(1)
concept representation
(1)
latent representation
(1)
language model
(1)
mechanistic interpretability
(1)
position bia
(1)
multilingual model
(1)
language bia
(1)
internal representation
(1)
multilingual transformer
(1)
data generation
(1)
multiple choice evaluation
(1)
activation patching
(1)
Papers
Hindsight Merging: Diverse Data Generation with Language Models
UAI 2025
Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers
ACL 2025
Do Llamas Work in English? On the Latent Language of Multilingual Transformers
ACL 2024
Evaluating Language Model Agency Through Negotiations
ICLR 2024
Self-Recognition in Language Models
EMNLP 2024