HKRAG: Holistic Knowledge Retrieval-Augmented Generation Over Visually-Rich Documents
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866918217836920832 |
|---|---|
| author | Tong, Anyang Niu, Xiang Liu, ZhiPing Tian, Chang Wei, Yanyan Shi, Zenglin Wang, Meng |
| author_facet | Tong, Anyang Niu, Xiang Liu, ZhiPing Tian, Chang Wei, Yanyan Shi, Zenglin Wang, Meng |
| contents | Existing multimodal Retrieval-Augmented Generation (RAG) methods for visually rich documents (VRD) are often biased towards retrieving salient knowledge(e.g., prominent text and visual elements), while largely neglecting the critical fine-print knowledge(e.g., small text, contextual details). This limitation leads to incomplete retrieval and compromises the generator's ability to produce accurate and comprehensive answers. To bridge this gap, we propose HKRAG, a new holistic RAG framework designed to explicitly capture and integrate both knowledge types. Our framework features two key components: (1) a Hybrid Masking-based Holistic Retriever that employs explicit masking strategies to separately model salient and fine-print knowledge, ensuring a query-relevant holistic information retrieval; and (2) an Uncertainty-guided Agentic Generator that dynamically assesses the uncertainty of initial answers and actively decides how to integrate the two distinct knowledge streams for optimal response generation. Extensive experiments on open-domain visual question answering benchmarks show that HKRAG consistently outperforms existing methods in both zero-shot and supervised settings, demonstrating the critical importance of holistic knowledge retrieval for VRD understanding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_20227 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | HKRAG: Holistic Knowledge Retrieval-Augmented Generation Over Visually-Rich Documents Tong, Anyang Niu, Xiang Liu, ZhiPing Tian, Chang Wei, Yanyan Shi, Zenglin Wang, Meng Information Retrieval Existing multimodal Retrieval-Augmented Generation (RAG) methods for visually rich documents (VRD) are often biased towards retrieving salient knowledge(e.g., prominent text and visual elements), while largely neglecting the critical fine-print knowledge(e.g., small text, contextual details). This limitation leads to incomplete retrieval and compromises the generator's ability to produce accurate and comprehensive answers. To bridge this gap, we propose HKRAG, a new holistic RAG framework designed to explicitly capture and integrate both knowledge types. Our framework features two key components: (1) a Hybrid Masking-based Holistic Retriever that employs explicit masking strategies to separately model salient and fine-print knowledge, ensuring a query-relevant holistic information retrieval; and (2) an Uncertainty-guided Agentic Generator that dynamically assesses the uncertainty of initial answers and actively decides how to integrate the two distinct knowledge streams for optimal response generation. Extensive experiments on open-domain visual question answering benchmarks show that HKRAG consistently outperforms existing methods in both zero-shot and supervised settings, demonstrating the critical importance of holistic knowledge retrieval for VRD understanding. |
| title | HKRAG: Holistic Knowledge Retrieval-Augmented Generation Over Visually-Rich Documents |
| topic | Information Retrieval |
| url | https://arxiv.org/abs/2511.20227 |