LiquidAI/ifstruct-v1.0
Dataset com 1 mil – 10 mil exemplos — 1 downloads no Hugging Face. IFStruct v1.0 Eval code: https://github.com/Liquid4All/ifstruct IFStruct is a benchmark for structured-output compliance: can a model produce valid JS…
O dataset LiquidAI/ifstruct-v1.0 está entre os destaques do Hugging Face — dados que alimentam o treinamento e a avaliação dos modelos do momento.
Ficha do dataset
- Tamanho: 1 mil – 10 mil exemplos
- Tarefas: geração de texto
- Formato/modalidade: JSON
- Idiomas: inglês
- Licença: Apache 2.0
- Downloads: 1 · Curtidas: 16
Sobre o dataset
IFStruct v1.0 Eval code: https://github.com/Liquid4All/ifstruct IFStruct is a benchmark for structured-output compliance: can a model produce valid JSON/YAML that follows a requested schema, when the requirements are phrased the many different ways real users phrase them? It is scored without constrained decoding, and only the structure is judged — not content quality, extraction accuracy, or reasoning — so the signal is isolated to output validity and schema following. Each… See the full description on the dataset page: https://huggingface.co/datasets/LiquidAI/ifstruct-v1.0.
Como carregar
Use a biblioteca datasets do Hugging Face:
pip install -U datasets
from datasets import load_dataset
ds = load_dataset("LiquidAI/ifstruct-v1.0")
print(ds)
print(ds["train"][0])Tags
text-generation structured-output json yaml instruction-following schema-following
Leia também
nvidia/Nemotron-Labs-TwoTower-30B-A3B-Base-BF16
OpenClaw is finally available on Android and iOS
Claude Science is Anthropic’s newest flagship product
