ClaimVer: Claim-Level Verification and Evidence Attribution Through Knowledge Graphs

ClaimVer is a dataset for explainable claim-level verification and evidence attribution of text, introduced in the arXiv preprint 'ClaimVer: Explainable Claim-Level Verification and Evidence Attribution of Text Through Knowledge Graphs'. The dataset was created by authors including Preetam Prabhu Srikar Dammu and uploaded to Hugging Face by user preetam7 on 2024-06-23.

Use Cases

• Training models for claim verification based on the described claim-level verification task.
• Developing methods for evidence attribution using the knowledge graph framework mentioned in the description.
• Researching explainable AI techniques for text analysis as suggested by the dataset's focus on explainability.

Strengths

• Dataset is associated with a detailed research paper (arXiv:2403.09724) providing methodology and context.
• Last updated on 2024-06-23, indicating recent maintenance.

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

Hugging Face user preetam7, associated with the research article 'ClaimVer: Explainable Claim-Level Verification and Evidence Attribution of Text Through Knowledge Graphs'.

Collection Method

Likely created for research on claim verification and knowledge graphs, as described in the linked arXiv paper.

Freshness

Last updated 2024-06-23 23:38:21.