Picture for Makoto Yamada

Makoto Yamada

Tokyo Institute of Technology

Many-to-Many Matching via Sparsity Controlled Optimal Transport

Add code
Mar 31, 2025
Viaarxiv icon

Efficient Model Editing with Task Vector Bases: A Theoretical Framework and Scalable Approach

Add code
Feb 03, 2025
Viaarxiv icon

On Verbalized Confidence Scores for LLMs

Add code
Dec 19, 2024
Figure 1 for On Verbalized Confidence Scores for LLMs
Figure 2 for On Verbalized Confidence Scores for LLMs
Figure 3 for On Verbalized Confidence Scores for LLMs
Figure 4 for On Verbalized Confidence Scores for LLMs
Viaarxiv icon

Learning Structured Representations with Hyperbolic Embeddings

Add code
Dec 02, 2024
Figure 1 for Learning Structured Representations with Hyperbolic Embeddings
Figure 2 for Learning Structured Representations with Hyperbolic Embeddings
Figure 3 for Learning Structured Representations with Hyperbolic Embeddings
Figure 4 for Learning Structured Representations with Hyperbolic Embeddings
Viaarxiv icon

Fast unsupervised ground metric learning with tree-Wasserstein distance

Add code
Nov 11, 2024
Viaarxiv icon

Towards the Effect of Examples on In-Context Learning: A Theoretical Case Study

Add code
Oct 12, 2024
Figure 1 for Towards the Effect of Examples on In-Context Learning: A Theoretical Case Study
Figure 2 for Towards the Effect of Examples on In-Context Learning: A Theoretical Case Study
Figure 3 for Towards the Effect of Examples on In-Context Learning: A Theoretical Case Study
Figure 4 for Towards the Effect of Examples on In-Context Learning: A Theoretical Case Study
Viaarxiv icon

Learning Structured Representations by Embedding Class Hierarchy with Fast Optimal Transport

Add code
Oct 04, 2024
Viaarxiv icon

Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis

Add code
Jun 16, 2024
Figure 1 for Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis
Figure 2 for Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis
Figure 3 for Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis
Figure 4 for Towards Understanding Jailbreak Attacks in LLMs: A Representation Space Analysis
Viaarxiv icon

Polyak Meets Parameter-free Clipped Gradient Descent

Add code
May 23, 2024
Viaarxiv icon

PhiNets: Brain-inspired Non-contrastive Learning Based on Temporal Prediction Hypothesis

Add code
May 23, 2024
Viaarxiv icon