Ekdeep Singh Lubana
~Ekdeep_Singh_Lubana1
18
论文总数
9.0
年均投稿
平均评分
接收情况15/18
会议分布
ICLR
9
NeurIPS
7
ICML
2
发表论文 (18 篇)
202511 篇
4
A Percolation Model of Emergence: Analyzing Transformers Trained on a Formal Language
ICLR 2025Poster
4
Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models
ICML 2025Poster
4
Projecting Assumptions: The Duality Between Sparse Autoencoders and Concept Geometry
NeurIPS 2025Poster
4
In-Context Learning Strategies Emerge Rationally
NeurIPS 2025Poster
4
Competition Dynamics Shape Algorithmic Phases of In-Context Learning
ICLR 2025Spotlight
5
From Flat to Hierarchical: Extracting Sparse Representations with Matching Pursuit
NeurIPS 2025Poster
4
Swing-by Dynamics in Concept Learning and Compositional Generalization
ICLR 2025Poster
4
ICLR: In-Context Learning of Representations
ICLR 2025Poster
4
Representation Shattering in Transformers: A Synthetic Study with Knowledge Editing
ICML 2025Poster
5
Representation Shattering in Transformers: A Synthetic Study with Knowledge Editing
ICLR 2025Rejected
5
Detecting High-Stakes Interactions with Activation Probes
NeurIPS 2025Poster
20247 篇
4
In-Context Learning Dynamics with Random Binary Sequences
ICLR 2024Poster
5
Abrupt Learning in Transformers: A Case Study on Matrix Completion
NeurIPS 2024Poster
4
What Makes and Breaks Safety Fine-tuning? A Mechanistic Study
NeurIPS 2024Poster
3
Mechanistically analyzing the effects of fine-tuning on procedurally defined tasks
ICLR 2024Poster
4
Emergence of Hidden Capabilities: Exploring Learning Dynamics in Concept Space
NeurIPS 2024Spotlight
4
How Capable Can a Transformer Become? A Study on Synthetic, Interpretable Tasks
ICLR 2024Rejected
4
Toward a Mechanistic Understanding of Stepwise Inference in Transformers: A Synthetic Graph Navigation Model
ICLR 2024Rejected