Hakaze Cho
Scholar

Hakaze Cho

Google Scholar ID: q_eQAcwAAAAJ
PhD Student, Japan Advanced Institute of Science and Technology
Machine LearningInterpretabilityIn-context Learning
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
43
 
H-index
4
 
i10-index
1
 
Publications
19
 
Co-authors
10
list available
Contact
No contact links provided.
Publications
11 items
Task Vectors, Learned Not Extracted: Performance Gains and Mechanistic Insight
2025
Cited
0
Localizing Task Recognition and Task Learning in In-Context Learning via Attention Head Analysis
2025
Cited
0
Mechanism of Task-oriented Information Removal in In-context Learning
2025
Cited
0
Binary Autoencoder for Mechanistic Interpretability of Large Language Models
2025
Cited
0
Unifying Attention Heads and Task Vectors via Hidden State Geometry in In-Context Learning
2025
Cited
0
Mechanistic Fine-tuning for In-context Learning
2025
Cited
0
Affinity and Diversity: A Unified Metric for Demonstration Selection via Internal Representations
2025
Cited
0
StaICC: Standardized Evaluation for Classification Task in In-context Learning
2025
Cited
0
Resume (English only)
Co-authors
10 total
Naoya Inoue
Naoya Inoue
Japan Advanced Institute of Science and Technology (JAIST) / RIKEN AIP
Mariko Kato
Mariko Kato
Japan Advanced Institute of Science and Technology
Haolin Yang
Haolin Yang
University of Chicago
Tetsushi Ohki
Tetsushi Ohki
College of Informatics, Shizuoka University
Yiqiao Zhong
Yiqiao Zhong
Assistant Professor, University of Wisconsin--Madison
Gouki Minegishi
Gouki Minegishi
University of Tokyo
Kaize Ding
Kaize Ding
Assistant Professor of Stats & Data Science, Northwestern University
Kentaro Inui
Kentaro Inui
MBZUAI, Tohoku University, RIKEN