REAL-MM-RAG: A Real-World Multi-Modal Retrieval Benchmark
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_ | 1866929718834495488 |
|---|---|
| author | Wasserman, Navve Pony, Roi Naparstek, Oshri Goldfarb, Adi Raz Schwartz, Eli Barzelay, Udi Karlinsky, Leonid |
| author_facet | Wasserman, Navve Pony, Roi Naparstek, Oshri Goldfarb, Adi Raz Schwartz, Eli Barzelay, Udi Karlinsky, Leonid |
| contents | Accurate multi-modal document retrieval is crucial for Retrieval-Augmented Generation (RAG), yet existing benchmarks do not fully capture real-world challenges with their current design. We introduce REAL-MM-RAG, an automatically generated benchmark designed to address four key properties essential for real-world retrieval: (i) multi-modal documents, (ii) enhanced difficulty, (iii) Realistic-RAG queries and (iv) accurate labeling. Additionally, we propose a multi-difficulty-level scheme based on query rephrasing to evaluate models' semantic understanding beyond keyword matching. Our benchmark reveals significant model weaknesses, particularly in handling table-heavy documents and robustness to query rephrasing. To mitigate these shortcomings, we curate a rephrased training set and introduce a new finance-focused, table-heavy dataset. Fine-tuning on these datasets enables models to achieve state-of-the-art retrieval performance on REAL-MM-RAG benchmark. Our work offers a better way to evaluate and improve retrieval in multi-modal RAG systems while also providing training data and models that address current limitations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_12342 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | REAL-MM-RAG: A Real-World Multi-Modal Retrieval Benchmark Wasserman, Navve Pony, Roi Naparstek, Oshri Goldfarb, Adi Raz Schwartz, Eli Barzelay, Udi Karlinsky, Leonid Information Retrieval Computer Vision and Pattern Recognition Accurate multi-modal document retrieval is crucial for Retrieval-Augmented Generation (RAG), yet existing benchmarks do not fully capture real-world challenges with their current design. We introduce REAL-MM-RAG, an automatically generated benchmark designed to address four key properties essential for real-world retrieval: (i) multi-modal documents, (ii) enhanced difficulty, (iii) Realistic-RAG queries and (iv) accurate labeling. Additionally, we propose a multi-difficulty-level scheme based on query rephrasing to evaluate models' semantic understanding beyond keyword matching. Our benchmark reveals significant model weaknesses, particularly in handling table-heavy documents and robustness to query rephrasing. To mitigate these shortcomings, we curate a rephrased training set and introduce a new finance-focused, table-heavy dataset. Fine-tuning on these datasets enables models to achieve state-of-the-art retrieval performance on REAL-MM-RAG benchmark. Our work offers a better way to evaluate and improve retrieval in multi-modal RAG systems while also providing training data and models that address current limitations. |
| title | REAL-MM-RAG: A Real-World Multi-Modal Retrieval Benchmark |
| topic | Information Retrieval Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.12342 |