Document Attribution: Examining Citation Relationships using Large Language Models
Research paper by Vipula Rawte and collaborators. Published 2025-05-09. Topics: Large Language Models (LLMs), document summarization, question answering, information extraction, attribution.
Resource information
- Authors
- Vipula Rawte, Ryan A. Rossi, Franck Dernoncourt, Nedim Lipka
- Research area
- Artificial intelligence
- Application domain
- General AI research
- AI methods
- Large language models
- Research topics
- cs.IR, cs.AI
- arXiv record date
- 2025-05-09
- Publication date
- 2025-05-09
- Publication
- arXiv
- Upstream last modified
- May 13, 2025
- License
- CC BY 4.0
- Last record update
- Nov 3, 2026
Abstract
As Large Language Models (LLMs) are increasingly applied to document-based tasks - such as document summarization, question answering, and information extraction - where user requirements focus on retrieving information from provided documents rather than relying on the model's parametric knowledge, ensuring the trustworthiness and interpretability of these systems has become a critical concern. A central approach to addressing this challenge is attribution, which involves tracing the generated outputs back to their source documents. However, since LLMs can produce inaccurate or imprecise responses, it is crucial to assess the reliability of these citations. To tackle this, our work proposes two techniques. (1) A zero-shot approach that frames attribution as a straightforward textual entailment task. Our method using flan-ul2 demonstrates an improvement of 0.27% and 2.4% over the best baseline of ID and OOD sets of AttributionBench, respectively. (2) We also explore the role of the attention mechanism in enhancing the attribution process. Using a smaller LLM, flan-t5-small, the F1 scores outperform the baseline across almost all layers except layer 4 and layers 8 through 11.