DocRetriever: A Plug-and-Play Framework for Multimodal Document Retrieval with Comprehensive Benchmark
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866918529517748224 |
|---|---|
| author | Hu, Ruofan Zhu, Menghui Zhu, Jieming Chen, Bo Xu, Shengyang Hong, Minjie Yang, Xiaoda Zhou, Sashuai Tang, Li Jin, Tao Zhao, Zhou |
| author_facet | Hu, Ruofan Zhu, Menghui Zhu, Jieming Chen, Bo Xu, Shengyang Hong, Minjie Yang, Xiaoda Zhou, Sashuai Tang, Li Jin, Tao Zhao, Zhou |
| contents | Multimodal documents contain diverse elements, such as tables, figures, and layouts, which can complicate retrieval tasks. While current approaches typically combine dense visual embedding models with supervised rerankers to achieve high-precision retrieval, they face inherent limitations. First, the coarse-grained nature of dense embeddings tends to obfuscate explicit semantics, failing to leverage structurally salient information. Second, supervised reranking models suffer from generalization bottlenecks, as their performance heavily relies on domain-specific training data. Furthermore, existing benchmarks often lack diverse assessment dimensions and comprehensive relevance annotations, limiting reliable evaluation. To address these challenges, we propose DocRetriever, a plug-and-play framework. It enhances visual retrieval via a layout-aware sparse embedding technique, enabling effective hybrid encoding without the overhead of optical character recognition (OCR). We also introduce a generalizable reranker that leverages reasoning-augmented demonstrations and optimized sampling to improve accuracy in few-shot settings. Finally, we construct a new benchmark, MultiDocR, to enable more rigorous evaluation. Experiments across diverse benchmarks validate DocRetriever's superiority over state-of-the-art methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_30027 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | DocRetriever: A Plug-and-Play Framework for Multimodal Document Retrieval with Comprehensive Benchmark Hu, Ruofan Zhu, Menghui Zhu, Jieming Chen, Bo Xu, Shengyang Hong, Minjie Yang, Xiaoda Zhou, Sashuai Tang, Li Jin, Tao Zhao, Zhou Computer Vision and Pattern Recognition Information Retrieval H.3.3; I.2.10 Multimodal documents contain diverse elements, such as tables, figures, and layouts, which can complicate retrieval tasks. While current approaches typically combine dense visual embedding models with supervised rerankers to achieve high-precision retrieval, they face inherent limitations. First, the coarse-grained nature of dense embeddings tends to obfuscate explicit semantics, failing to leverage structurally salient information. Second, supervised reranking models suffer from generalization bottlenecks, as their performance heavily relies on domain-specific training data. Furthermore, existing benchmarks often lack diverse assessment dimensions and comprehensive relevance annotations, limiting reliable evaluation. To address these challenges, we propose DocRetriever, a plug-and-play framework. It enhances visual retrieval via a layout-aware sparse embedding technique, enabling effective hybrid encoding without the overhead of optical character recognition (OCR). We also introduce a generalizable reranker that leverages reasoning-augmented demonstrations and optimized sampling to improve accuracy in few-shot settings. Finally, we construct a new benchmark, MultiDocR, to enable more rigorous evaluation. Experiments across diverse benchmarks validate DocRetriever's superiority over state-of-the-art methods. |
| title | DocRetriever: A Plug-and-Play Framework for Multimodal Document Retrieval with Comprehensive Benchmark |
| topic | Computer Vision and Pattern Recognition Information Retrieval H.3.3; I.2.10 |
| url | https://arxiv.org/abs/2605.30027 |