← Search

Takumi Okamoto

2 accepted papers

2026

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

ICLR 2026oral

Empirical scaling laws have driven the evolution of large language models (LLMs), yet their coefficients shift whenever the model architecture or data pipeline changes. Mixture‑of‑Experts (MoE) models, now standard in state‑of‑the‑art systems, introduce a new sparsity dimension that current dense‑mo…

Cited by 0SourcecodeScholar
2026

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

ICLR 2026poster

The performance of large language models (LLMs) in program synthesis and mathematical reasoning is fundamentally limited by the quality of their pre-training corpora. We introduce two openly licensed pre-training datasets, released under the Llama 3.3 Community License, that significantly enhance…

Cited by 0SourcecodeScholar