Learning to Refine with Fine-Grained Natural Language Feedback
Paper • 2407.02397 • Published
Model Card for Model ID
Model Details
Given a document and a factually inconsistent claim, this model generates an explanation for why the claim is inconsistent with the document along with fine-grained span with the inconsistency.
Model Sources
- Repository: https://github.com/ManyaWadhwa/DCR
- Paper: https://arxiv.org/pdf/2407.02397
- Downloads last month
- 1
Model tree for wadhma/Critique-L3-FT-DCR
Base model
meta-llama/Meta-Llama-3-8B