Taishi

Taishi-N324's activity

• All
• Models
• Datasets
• Spaces
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

authored a paper 5 days ago

On the Optimal Reasoning Length for RL-Trained Language Models

Paper • 2602.09591 • Published 6 days ago • 3

authored a paper 5 months ago

MixtureVitae: Open Web-Scale Pretraining Dataset With High Quality Instruction and Reasoning Data Built from Permissive-First Text Sources

Paper • 2509.25531 • Published Sep 29, 2025 • 9

authored a paper 6 months ago

Optimal Sparsity of Mixture-of-Experts Language Models for Reasoning Tasks

Paper • 2508.18672 • Published Aug 26, 2025 • 10

authored a paper 8 months ago

Rewriting Pre-Training Data Boosts LLM Performance in Math and Code

Paper • 2505.02881 • Published May 5, 2025 • 6

authored a paper 11 months ago

Building Instruction-Tuning Datasets from Human-Written Instructions with Open-Weight Large Language Models

Paper • 2503.23714 • Published Mar 31, 2025 • 1

authored 4 papers 12 months ago

Balancing Speed and Stability: The Trade-offs of FP8 vs. BF16 Training in LLMs

Paper • 2411.08719 • Published Nov 10, 2024 • 1

Why We Build Local Large Language Models: An Observational Analysis from 35 Japanese and Multilingual LLMs

Paper • 2412.14471 • Published Dec 19, 2024

Wider or Deeper? Scaling LLM Inference-Time Compute with Adaptive Branching Tree Search

Paper • 2503.04412 • Published Mar 6, 2025 • 5

Drop-Upcycling: Training Sparse Mixture of Experts with Partial Re-initialization

Paper • 2502.19261 • Published Feb 26, 2025 • 6

authored 4 papers over 1 year ago

Agent Skill Acquisition for Large Language Models via CycleQD

Paper • 2410.14735 • Published Oct 16, 2024 • 2

Continual Pre-Training for Cross-Lingual LLM Adaptation: Enhancing Japanese Language Capabilities

Paper • 2404.17790 • Published Apr 27, 2024 • 5

Building a Large Japanese Web Corpus for Large Language Models

Paper • 2404.17733 • Published Apr 27, 2024 • 5

LLM-jp: A Cross-organizational Project for the Research and Development of Fully Open Japanese LLMs

Paper • 2407.03963 • Published Jul 4, 2024 • 18

authored a paper almost 2 years ago

Aurora-M: The First Open Source Multilingual Language Model Red-teamed according to the U.S. Executive Order

Paper • 2404.00399 • Published Mar 30, 2024 • 42