datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
2WikiMultihopQAMirror of https://github.com/Alab-NII/2wikimultihoptvs-2wikimultihopqa
TVS-2WikiMultiHopQA Dataset
This repository contains the tvs-2wikimultihopqa dataset, which is a key component of the "Think, Verbalize, then Speak" (TVS) framework presented in the paper Think, Verbalize, then Speak: Bridging Complex Thoughts and Comprehensible Speech.
Project Page: https://yhytoto12.github.io/TVS-ReVerT
Paper: https://huggingface.co/papers/2509.16028
Code: https://github.com/yhytoto12/TVS-ReVerT
Introduction
The "Think, Verbalize, then Speak"… See the full description on the dataset page: https://huggingface.co/datasets/yhytoto12/tvs-2wikimultihopqa.adaptive_rag_2wikimultihopqaIn this collection you can find 4 datasets with is_supporting=True contexts from the Adaptive RAG collection.
There are picked 4/6 datasets from Adaptive RAG datasets with is_supporting=True contexts.
Not all samples from TriviaQA and SQUAD have is_supporting=True contexts, thats why we do not include them in hf collection.
Script for data transformation from original Adaptive RAG format into our format can be found here:… See the full description on the dataset page: https://huggingface.co/datasets/aboriskin/adaptive_rag_2wikimultihopqa.