Hiroshi Matsuda

2 paper cards in Mova's library · 2025–2026

Hiroshi Matsuda wrote 2 of the papers Mova reads, in 2025–2026, most of them on Dependency parsing, Treebanks; often with Masayuki Asahara, Chunpeng Ma.

The profile is the library's match of the name; two people with one name are not always told apart.

Papers

Probing the Dynamics of Syntactic Ability Acquisition Throughout LLM Pretraining (2026)
The authors study when syntactic skills appear during LLM pretraining and which ones stay weak. Their method, LoRA probing, keeps the base model frozen and trains only low-rank adapters to produce step-by-step UD parses in a simple tabular format. They apply it to the public intermediate checkpoints of OLMo-2-7B…
Step-by-step Instructions and a Simple Tabular Output Format Improve the Dependency Parsing Accuracy of LLMs (2025)
The paper asks how to get structurally valid, accurate Universal Dependencies parses out of large language models.