Ekdeep Singh Lubana
18 papers · 2021–2026 · 5 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 ↓+6 more ↓ Show less ↑
πΊοΈ Taxonomy Completionist (32) π Academic Marathon Β· Sprint π Conference Polyglot Β· Silver π Interdisciplinary Bridge Β· Silver π£ Hot Topic Early Bird
π
Triple Crown
π€
Dynamic Duo
(12)
β‘
Prolific Year Β· Productive
π
Century Club Β· Bronze
ποΈ
Keyword Collector
(90)
β
The Questioner
(3)
Venues
Top co-authors
Research topics
Machine Learning
(15)
Artificial Intelligence
(11)
Deep Learning
(11)
Core AI
(11)
Optimization & Theory Β· Machine Learning
(10)
Core Methods
(7)
Architectures
(6)
Learning Types Β· Machine Learning
(5)
Natural Language Processing
(3)
Learning Types Β· Deep Learning
(3)
Models
(3)
Resources & Methods
(3)
Optimization & Theory Β· Deep Learning
(2)
Techniques
(2)
Interdisciplinary
(1)
Keywords
representation learning
(4)
learning dynamics
(3)
formal language
(3)
in-context learning
(3)
loss landscape
(2)
neural network optimization
(2)
neural network
(2)
mechanistic interpretability
(2)
compositional generalization
(2)
emergent capability
(2)
concept learning
(2)
transformer
(2)
sparse autoencoder
(2)
network pruning
(2)
energy minimization
(1)
autoregressive transformer
(1)
preference optimization
(1)
feature learning
(1)
preference alignment
(1)
knowledge editing
(1)
Papers
From Isolation to Entanglement: When Do Interpretability Methods Identify and Disentangle Known Concepts?
ACL 2026
Analyzing (In)Abilities of SAEs via Formal Languages
NAACL 2025
A Percolation Model of Emergence: Analyzing Transformers Trained on a Formal Language
ICLR 2025
Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models
ICML 2025
Competition Dynamics Shape Algorithmic Phases of In-Context Learning
ICLR 2025
ICLR: In-Context Learning of Representations
ICLR 2025
Representation Shattering in Transformers: A Synthetic Study with Knowledge Editing
ICML 2025
Swing-by Dynamics in Concept Learning and Compositional Generalization
ICLR 2025
Abrupt Learning in Transformers: A Case Study on Matrix Completion
NeurIPS 2024
Compositional Capabilities of Autoregressive Transformers: A Study on Synthetic, Interpretable Tasks
ICML 2024
Emergence of Hidden Capabilities: Exploring Learning Dynamics in Concept Space
NeurIPS 2024
In-Context Learning Dynamics with Random Binary Sequences
ICLR 2024
Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks
ICLR 2024
Towards an Understanding of Stepwise Inference in Transformers: A Synthetic Graph Navigation Model
ICML 2024
What Makes and Breaks Safety Fine-tuning? A Mechanistic Study
NeurIPS 2024
Mechanistic Mode Connectivity
ICML 2023
What shapes the loss landscape of self supervised learning?
ICLR 2023
A Gradient Flow Framework For Analyzing Network Pruning
ICLR 2021