arXiv Computer Vision

Invoice Haystack: Benchmarking Document Retrieval and Visual Question Answering Under Strong Visual Homogeneity

The paper introduces Invoice Haystack, a benchmark of 1,500 anonymized invoices and 200 question‑answer pairs that tests document retrieval and visual question answering under strong visual homogeneity. It shows that existing benchmarks suffer from embedding collapse, with Invoice Haystack’s mean pairwise cosine similarity at 0.73 versus 0.38 and 0.31 in DocHaystack and InfoHaystack. The authors propose VL‑RAG, a hybrid retrieval‑augmented generation framework that combines text and visual embeddings and a VLM‑based verification filter, achieving 60.0% Recall@1 on Invoice Haystack‑500 and improving performance on other benchmarks.

arXiv Computer Vision

Aug 27

VisDocAgentBench: Benchmarking Agents for Visually Rich Document Retrieval

VisDocAgentBench is a closed‑corpus benchmark that evaluates static versus agentic retrieval for visually rich documents, using 2,375 pages from 100 documents and 120 queries that span direct, one‑bridge, and two‑bridge evidence structures. The benchmark includes semantic, relational, and visual queries, full‑document review, and hard‑negative validation. Results show that a strong visual retriever performs well on direct items but poorly on two‑bridge items, while agents improve performance, especially when using visual retrieval and iterative search capabilities.

By Lexiang Hu, Yanzhao Zhang, Mingxin Li, Dingkun Long, Yikang Li, Fuwei Zhang, Yisen Wang, Zhouchen Lin

agents computer-vision nlp benchmarks

arXiv Computation and Language

3d ago

TestHallVQA: Exploring LVLMs' Document-Level Reasoning under Redundant Contexts from Scientific Exams

arXiv:2609.13158v1 Announce Type: new Abstract: Large Vision--Language Models (LVLMs) are increasingly expected to perform visual question answering (VQA) over planar media. However, existing planar...

By Yongqi Yu, Yu Zhang

llms nlp benchmarks

arXiv AI

Jul 29

VLD-RAG: Agentic Vision-Language Retrieval-Augmented Generation for Long, Visually-Rich Multi-Page Documents

arXiv:2607. 24748v1 Announce Type: cross Abstract: Visually-rich documents such as reports, slides, and manuals often distribute the evidence needed to answer a question across multiple pages, mixing text with layout cues, tables, charts, and figures.

By Seonok Kim

rag agents nlp multimodal benchmarks

arXiv AI

Sep 3

ViSAR: Training-Free Adaptive-$k$ Retrieval for Visual Document Question Answering

ViSAR is a training‑free, adaptive‑k retrieval method for Visual Document Question Answering that operates directly in the embedding space to build a query‑conditioned page‑level similarity matrix. By dynamically selecting the number of pages to retrieve based on query relevance, ViSAR reduces Retrieval‑Augmented Generation latency by up to 58.7% while maintaining or improving answer accuracy across multiple encoders and Large Vision‑Language Models. The structure of the similarity matrix also correlates with answer accuracy, indicating potential for retrieval quality‑aware document understanding.

By Adrien Mialland, Marc Plantevit, Julien Gallois, C\'eline Robardet

llms rag nlp multimodal

arXiv AI

Sep 10

Bridging the Semantic-Utility Gap in Multimodal RAG via Generator-in-the-Loop Alignment

arXiv:2609.08188v1 Announce Type: new Abstract: Vision-language models (VLMs) augmented with retrieval-augmented generation (RAG) benefit from access to external evidence. However, standard retriever...

By Zhan-Lun Chang, Dong-Jun Han, Seyyedali Hosseinalipour, Mung Chiang, Christopher G. Brinton

llms rag reinforcement-learning fine-tuning multimodal safety

arXiv Computer Vision

Sep 1

Doc-REFRAG: Rethinking Multimodal Document Retrieval-Augmented Generation

arXiv:2608.30163v1 Announce Type: cross Abstract: Real-world knowledge resides in multimodal documents, necessitating retrieval-augmented generation (RAG) for accurate question answering. However, ex...

By Ruofan Hu, Shengyang Xu, Minjie Hong, Xiaoda Yang, Sashuai Zhou, Ke Lei, Tao Jin, Zhou Zhao

rag nlp reinforcement-learning efficiency multimodal benchmarks

arXiv Machine Learning

Jul 9

Comparative Study of Domain-adapted VLMs for General Document Visual Question Answering

arXiv:2607. 07179v1 Announce Type: cross Abstract: Document Visual Question Answering (DocVQA) presents a complex multimodal challenge, requiring models to exploit visual, textual, and layout information from documents.

By Miguel Lopez-Duran, Elena Marrero, Julian Fierrez, Marta Robledo-Moreno, Ruben Vera-Rodriguez, Daniel DeAlcala, Aythami Morales, Ruben Tolosana, Oscar Delgado, Alvaro Ortigosa, Javier Ortega-Garcia

llms nlp fine-tuning multimodal

arXiv AI

Jul 29

Multimodal Hybrid Retrieval-Augmented Generation for Scientific Document Understanding using Open-Source SLMs

arXiv:2607. 24799v1 Announce Type: cross Abstract: Large Language Models tend to hallucinate when answering domain-specific ques tions from scientific documents without prior fine-tuning.

By Alexandru-Andrei Sauc\u{a}, Ana-Luiza Rusnac

llms rag fine-tuning multimodal benchmarks

arXiv Machine Learning

Jul 24

Multimodal CoLRAG-TF: Triple-Filtered Retrieval for Complex PDFs

arXiv:2607. 20517v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) over heterogeneous PDF collections remains challenging due to multimodal content, domain-specific terminology, and the need for multi-hop reasoning across dispersed evidence.

By Takato Yasuno

llms rag computer-vision multimodal benchmarks safety

arXiv Computer Vision

Aug 24

Generative Embedding Benchmark: How Much Information Survives in a Dense Embedding?

The paper introduces the Generative Embedding Benchmark (GEB), which evaluates how much content from an embedding can be recovered by a decoder that only has access to the frozen embedding and a question, without the original image or intermediate features. GEB uses a curated visual‑question‑answering dataset with 1,800 development and 900 test items covering natural images, scene text, and visual documents. Experiments on seven public embedding models show that visual‑only scores range from 28.25 to 33.21, while joint image‑question encoding boosts scores up to 65.56, revealing that generative readout uncovers information bottlenecks not captured by traditional separability‑based benchmarks.

By Yun Li, Biao Yang, Peixi Wu, Yunhao Zhou, Mingzhou Jiang, Wei Yuan, Fan Yang, Wenwu Ou

rag multimodal benchmarks

arXiv Computer Vision

Aug 25

Beyond Visual Similarity: Entity-Aligned Retrieval for Knowledge-Based Visual Question Answering

arXiv:2608.21450v1 Announce Type: new Abstract: Knowledge-Based Visual Question Answering (KB-VQA) relies on retrieving external information to answer queries involving long-tail entities. However, e...

By Hangrui Xu, Zhengxian Wu, Yunyao Yu, Zhuohong Chen, Rui Cong, Xiangwen Deng, Zhifang Liu, Peng Jiao, Haoqian Wang

rag nlp efficiency safety

arXiv AI

Jun 30

PIXELRAG: Web Screenshots Beat Text for Retrieval-Augmented Generation

arXiv:2606. 28344v1 Announce Type: cross Abstract: Augmenting large language models (LLMs) with retrieved web text has become a dominant paradigm, yet the web is not natively textual: existing systems depend on complex parsing pipelines that linearize HTML and discard layout, visual structure, and formatting.

By Yichuan Wang, Zhifei Li, Zirui Wang, Paul Teiletche, Lesheng Jin, Matei Zaharia, Joseph E. Gonzalez, Sewon Min

llms rag agents multimodal benchmarks

Related stories