Skip to content

Dataset · RUCKBReasoning

Table-based SFT training data

RUCKBReasoning/TableLLM-SFT

English training dataset for fine-tuning language models on table question answering.

What's inside

The dataset is an English text training set combining multiple splits from various tabular benchmarks. The authors used this data for SFT fine-tuning of the TableLLM-8b model based on Llama 3.1 8B Instruct.

What you can do with it

  • SFT fine-tuning of LLMs and LoRA adapters for tabular data analysis
  • Building AI assistants for question answering on reports and tables
  • Comparing tabular reasoning quality across combined benchmarks
Tasks
Table question answering, Text generation
Languages
English
Downloads this month
715
Likes
30

Models trained on it

How to load it

from datasets import load_dataset

ds = load_dataset("RUCKBReasoning/TableLLM-SFT")