# Document Attribution: Examining Citation Relationships using Large Language Models

arXiv: https://arxiv.org/abs/2505.06324 · HTML: https://mlsanity.com/paper/2505.06324/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 12 references that are in the map.
- First posted on arXiv 9 May 2025.

## Builds on (most cited first, 7 of 7)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer](https://mlsanity.com/paper/1910.10683/) (2019)
- [A Survey of Large Language Models Attribution](https://mlsanity.com/paper/2311.03731/) (2023)
- [Using Captum to Explain Generative Language Models](https://mlsanity.com/paper/2312.05491/) (2023)
- [AttributionBench: How Hard is Automatic Attribution Evaluation?](https://mlsanity.com/paper/2402.15089/) (2024)
- [On the Limitations of Large Language Models (LLMs): False Attribution](https://mlsanity.com/paper/2404.04631/) (2024)
- [LLM Attributor: Interactive Visual Attribution for LLM Generation](https://mlsanity.com/paper/2404.01361/) (2024)

## Built on by (most cited first, 1 of 1)

- [Beyond Factual Accuracy: Evaluating Global Reasoning Integrity in RAG Systems with LogicScore](https://mlsanity.com/paper/2601.15050/) (2026)

## Abstract (arXiv)

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.