File access restricted

ClaimFlow: Tracing the Evolution of Scientific Claims in NLP

Files

ClaimFlow-v0.1.zip (434.35 KB)

Date

2026-06-24

Type

Dataset

Authors

Pramanick, Aniket
Hou, Yufang
Mohammad, Saif M.
Iryna, Gurevych

Description

ClaimFlow is a dataset designed to support the study of scientific claims and their relationships across research papers. ## Dataset Version * Dataset: ClaimFlow * Version: Preview v0.1 * Status: Unpublished Research Resource * Owner: Aniket Pramanick ## Contents * `data/sample_gold_small.jsonl` — Preview dataset. * `metadata/schema.json` — Description of dataset fields. * `metadata/dataset_info.json` — Dataset metadata. ## Usage Restrictions This dataset is shared exclusively for research and evaluation purposes. By accessing or using this dataset, you agree that: 1. You will not redistribute, publish, or share the dataset with third parties. 2. You will not use the dataset in any publication, benchmark, product, or public release without prior written permission from the dataset owner. 3. You will not claim ownership of the dataset or any derivative resource based on it. 4. You will provide appropriate citation and acknowledgment when the associated ClaimFlow paper becomes available. 5. You understand that the dataset remains unpublished work. ## Feedback Feedback, comments, and suggestions are welcome and appreciated. For questions regarding the dataset, please contact the authors: [Aniket Pramanick](mailto:aniketpramanick26@gmail.com), [Yufang Hou](mailto:yufang.hou@it-u.at), [Saif M. Mohammad](saif.mohammad@nrc-cnrc.gc.ca). ## Limitations and Disclaimer This dataset is provided for research purposes only. While considerable effort has been made to ensure the quality of the annotations and metadata, the dataset may contain errors, omissions, inconsistencies, annotation mistakes, parsing errors, or incomplete records. Users are responsible for validating the data before drawing conclusions or using it in downstream analyses. The dataset owner makes no guarantees regarding the accuracy, completeness, or suitability of the dataset for any particular purpose. The dataset is provided "as is" without warranty of any kind. The owner makes no guarantees regarding completeness, correctness, or fitness for a particular purpose. Copyright © Ubiquitous Knowledge Processing (UKP) Lab, Technische Universität Darmstadt. All rights reserved.