# CiteVQA: Benchmarking Evidence Attribution for Trustworthy Document Intelligence

arXiv: https://arxiv.org/abs/2605.12882 · HTML: https://mlsanity.com/paper/2605.12882/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Builds on 52 of its 84 references that are in the map.
- First posted on arXiv 13 May 2026.

## Builds on (most cited first, 30 of 52)

- [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)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [WebGPT: Browser-assisted question-answering with human feedback](https://mlsanity.com/paper/2112.09332/) (2021)
- [LAION-400M: Open Dataset of CLIP-Filtered 400 Million Image-Text Pairs](https://mlsanity.com/paper/2111.02114/) (2021)
- [ChartQA: A Benchmark for Question Answering about Charts with Visual and Logical Reasoning](https://mlsanity.com/paper/2203.10244/) (2022)
- [What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams](https://mlsanity.com/paper/2009.13081/) (2020)
- [DocVQA: A Dataset for VQA on Document Images](https://mlsanity.com/paper/2007.00398/) (2020)
- [PubMedQA: A Dataset for Biomedical Research Question Answering](https://mlsanity.com/paper/1909.06146/) (2019)
- [FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generation](https://mlsanity.com/paper/2305.14251/) (2023)
- [Kosmos-2: Grounding Multimodal Large Language Models to the World](https://mlsanity.com/paper/2306.14824/) (2023)
- [Gemma: Open Models Based on Gemini Research and Technology](https://mlsanity.com/paper/2403.08295/) (2024)
- [Enabling Large Language Models to Generate Text with Citations](https://mlsanity.com/paper/2305.14627/) (2023)
- [GLM-5: from Vibe Coding to Agentic Engineering](https://mlsanity.com/paper/2602.15763/) (2026)
- [Ferret: Refer and Ground Anything Anywhere at Any Granularity](https://mlsanity.com/paper/2310.07704/) (2023)
- [LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking](https://mlsanity.com/paper/2204.08387/) (2022)
- [DeepEyes: Incentivizing "Thinking with Images" via Reinforcement Learning](https://mlsanity.com/paper/2505.14362/) (2025)
- [OCR-free Document Understanding Transformer](https://mlsanity.com/paper/2111.15664/) (2021)
- [Retrieval-Augmented Generation for AI-Generated Content: A Survey](https://mlsanity.com/paper/2402.19473/) (2024)
- [VisRAG: Vision-based Retrieval-augmented Generation on Multi-modality Documents](https://mlsanity.com/paper/2410.10594/) (2024)
- [CharXiv: Charting Gaps in Realistic Chart Understanding in Multimodal LLMs](https://mlsanity.com/paper/2406.18521/) (2024)
- [WebSailor: Navigating Super-human Reasoning for Web Agent](https://mlsanity.com/paper/2507.02592/) (2025)
- [mPLUG-DocOwl 1.5: Unified Structure Learning for OCR-free Document Understanding](https://mlsanity.com/paper/2403.12895/) (2024)
- [Multimodal C4: An Open, Billion-scale Corpus of Images Interleaved with Text](https://mlsanity.com/paper/2304.06939/) (2023)
- [ColPali: Efficient Document Retrieval with Vision Language Models](https://mlsanity.com/paper/2407.01449/) (2024)
- [MMLongBench-Doc: Benchmarking Long-context Document Understanding with Visualizations](https://mlsanity.com/paper/2407.01523/) (2024)
- [SlideVQA: A Dataset for Document Visual Question Answering on Multiple Images](https://mlsanity.com/paper/2301.04883/) (2023)
- [Thyme: Think Beyond Images](https://mlsanity.com/paper/2508.11630/) (2025)
- [OmniDocBench: Benchmarking Diverse PDF Document Parsing with Comprehensive Annotations](https://mlsanity.com/paper/2412.07626/) (2024)
- [Document Understanding Dataset and Evaluation (DUDE)](https://mlsanity.com/paper/2305.08455/) (2023)

## Built on by (most cited first, 5 of 5)

- [Beyond Frame Selection: Generative Latent Evidence Aggregation for Long-Video Understanding](https://mlsanity.com/paper/2607.28516/) (2026)
- [MedBench v5: A Dynamic, Process-Oriented, and Hallucination-Aware Benchmark for Clinical Multimodal Models](https://mlsanity.com/paper/2606.24155/) (2026)
- [Evidence Attribution in Visual Document Understanding without Coordinates or Region Labels](https://mlsanity.com/paper/2607.24651/) (2026)
- [DocAttriBench: Benchmarking Answer Grounding in Document Visual Question Answering](https://mlsanity.com/paper/2609.20574/) (2026)
- [Do Coding Agents Reuse Existing Code or Reinvent the Wheel?](https://mlsanity.com/paper/2609.35357/) (2026)

## Abstract (arXiv)

Multimodal Large Language Models (MLLMs) have significantly advanced document understanding, yet current Doc-VQA evaluations score only the final answer and leave the supporting evidence unchecked. This answer-only approach masks a critical failure mode: a model can land on the correct answer while grounding it in the wrong passage---a critical risk in high-stakes domains like law, finance, and medicine, where every conclusion must be traceable to a specific source region. To address this, we introduce CiteVQA, a benchmark that requires models to return textitelement-level bounding-box citations alongside each answer, evaluating both jointly. CiteVQA comprises 1,897 questions across 711 PDFs spanning seven domains and two languages, averaging 40.6 pages per document. To ensure fidelity and scalability, the ground-truth citations are generated by an automated pipeline---which identifies crucial evidence via masking ablation and enforces multi-stage quality control. At the core of our evaluation is Strict Attributed Accuracy (SAA), which credits a prediction only when the answer and the cited region are both correct. Auditing 20 MLLMs reveals a pervasive Attribution Hallucination: models frequently produce the right answer while citing the wrong region. The strongest system (Gemini-3.1-Pro-Preview) achieves an SAA of only 76.0, and the strongest open-source MLLM reaches just 22.5. Ultimately, towards trustworthy document intelligence, CiteVQA exposes a reliability gap that answer-only evaluations overlook, providing the instrumentation needed to close it. Our repository is available at https://github.com/opendatalab/CiteVQA.