CoUDA: Coherence Evaluation via Unified Data Augmentation

[Free GPT-4]
[PDF] arxiv.org

From Single to Multi: How LLMs Hallucinate in Multi-Document Summarization

CG Belem, P Pezeskhpour, H Iso, S Maekawa… - arxiv preprint arxiv …, 2024 - arxiv.org
Although many studies have investigated and reduced hallucinations in large language
models (LLMs) for single-document tasks, research on hallucination in multi-document …
Save Cite Related articles View as HTML