Maritime SFT Mixed Formats: Domain-Specific Training Data

A supervised fine-tuning dataset for the maritime domain generated from maritime books and technical documents using the GLM-4 model via NVIDIA NIM API. The dataset is provided in three distinct formats: Alpaca, ChatML, and RAG triad. It was created by author uday568 and last updated on May 14, 2026.

Use Cases

• Fine-tune language models for maritime question-answering based on the described RAG triad format.
• Train conversational AI assistants for maritime technical support based on the ChatML multi-turn format.
• Develop instruction-following models for maritime procedures based on the standard Alpaca SFT format.

Strengths

• Dataset is explicitly designed for supervised fine-tuning with high-quality generation noted.
• Provides data in three distinct structured formats (Alpaca, ChatML, RAG) for different training scenarios.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Freshness should be verified as the last update timestamp is in the future (2026-05-14).

Provenance

Source

Generated from maritime books and technical documents.

Collection Method

Generated using GLM-4 via NVIDIA NIM API.

Freshness

Last updated 2026-05-14 09:04:33