Gouki Minegishi
6 papers · 2024–2026 · 6 conferences · across top CS/AI conferences
Achievements
Jump to papers ↓
π
Conference Polyglot
(5)
π
Cross-Pollinator
(8)
π
Interdisciplinary Bridge
πΊοΈ
Taxonomy Completionist
(11)
π§
Keyword Pioneer
Conferences
AACL (1)
ACL (1)
ICLR (1)
ICML (1)
IJCNLP (1)
NIPS (1)
Top co-authors
Keywords
numerical reasoning
(2)
linear probing
(2)
representation learning
(1)
stochastic gradient descent
(1)
hyperparameter optimization
(1)
neural network optimization
(1)
learning rate
(1)
convergence guarantee
(1)
convergence rate
(1)
adam optimizer
(1)
adaptive gradient method
(1)
deep learning optimization
(1)
momentum method
(1)
mechanistic interpretability
(1)
sparse autoencoder
(1)
numerical attribute
(1)
representation analysis
(1)
multi-attribute confounding
(1)
large language model
(1)
deep learning optimizer
(1)
Papers
Understanding Emergent Misalignment via Feature Superposition Geometry
ACL 2026
Interpreting Multi-Attribute Confounding through Numerical Attributes in Large Language Models
AACL 2025
Rethinking Evaluation of Sparse Autoencoders through the Representation of Polysemous Words
ICLR 2025
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
ADOPT: Modified Adam Can Converge with Any $\beta_2$ with the Optimal Rate
NIPS 2024