Gouki Minegishi
5 papers · 2024–2026 · 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 ↓
🌍
Conference Polyglot · Silver
🌉
Interdisciplinary Bridge · Bronze
🗺️
Taxonomy Completionist
(14)
Venues
Top co-authors
Research topics
Keywords
sparse autoencoder
(2)
deep reinforcement learning
(1)
representation learning
(1)
stochastic gradient descent
(1)
hyperparameter optimization
(1)
semantic representation
(1)
neural network optimization
(1)
learning rate
(1)
convergence guarantee
(1)
convergence rate
(1)
adam optimizer
(1)
induction head
(1)
adaptive gradient method
(1)
deep learning optimization
(1)
momentum method
(1)
mechanistic interpretability
(1)
linear probing
(1)
numerical attribute
(1)
numerical reasoning
(1)
representation analysis
(1)
Papers
Understanding Emergent Misalignment via Feature Superposition Geometry
ACL 2026
Beyond Induction Heads: In-Context Meta Learning Induces Multi-Phase Circuit Emergence
ICML 2025
Interpreting Multi-Attribute Confounding through Numerical Attributes in Large Language Models
IJCNLP 2025
Rethinking Evaluation of Sparse Autoencoders through the Representation of Polysemous Words
ICLR 2025
ADOPT: Modified Adam Can Converge with Any β_2 with the Optimal Rate
NeurIPS 2024