Nemotron RL Instruction Following Free Form Formatting V1: Text Formatting Training Data

A dataset created by NVIDIA Corporation on April 10, 2026, designed to teach models to follow arbitrary text formatting instructions. It uses explicit Regex and string matching for the reward signal and is intended for commercial or non-commercial use.

Use Cases

• Training reinforcement learning agents for text formatting based on explicit reward signals.
• Fine-tuning language models to follow instructions for bullet styles and numbering.
• Improving model adherence to specific heading formats and inline emphasis styles.
• Developing models that can structure web-style answers based on formatting guidelines.

Strengths

• Created by a major AI research organization, NVIDIA Corporation.
• Explicitly designed for commercial and non-commercial use.
• Uses explicit Regex and string matching for a defined reward signal.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

NVIDIA Corporation

Collection Method

Likely generated for reinforcement learning training.

Time Range

Created April 10, 2026.

Freshness

Last updated 2026-06-04 05:10:45.

Geography

null