MARQUIS: A Three-Stage Pipeline for Video Retrieval-Augmented Generation
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_ | 1866917506186215424 |
|---|---|
| author | Chakraborty, Debashish Zhang, Dengjia Jin, Jialiang Liu, Hanting Guerrerio, Katherine Qin, Hanxiang Skow, Tyler Martin, Alexander Kriz, Reno Van Durme, Benjamin |
| author_facet | Chakraborty, Debashish Zhang, Dengjia Jin, Jialiang Liu, Hanting Guerrerio, Katherine Qin, Hanxiang Skow, Tyler Martin, Alexander Kriz, Reno Van Durme, Benjamin |
| contents | Retrieval-augmented generation from videos requires systems to retrieve relevant audiovisual evidence from large corpora and synthesize it into coherent, attributed text. Current approaches struggle at both ends: retrieval methods fail on complex, multi-faceted queries that cannot be captured by a single embedding, while generation methods lack the high-level reasoning needed to synthesize across multiple videos and face memory constraints over long, multi-video contexts. We present MARQUIS: a three-stage pipeline that addresses these limitations through (1) query expansion, fusion, and reranking, (2) calibrated structured evidence extraction, and (3) article generation from extracted evidence, optionally controlled by an RLM. On the MAGMaR2026 shared task, we improve retrieval performance from 0.195 to 0.759 (nDCG@10). For article generation, ITER-QA-BASE improves average human score from 3.09 to 3.83 over the CAG baseline, while MARQUIS-RLM achieves a human score of 3.30 and the strongest citation recall among non-QA systems. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_17640 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | MARQUIS: A Three-Stage Pipeline for Video Retrieval-Augmented Generation Chakraborty, Debashish Zhang, Dengjia Jin, Jialiang Liu, Hanting Guerrerio, Katherine Qin, Hanxiang Skow, Tyler Martin, Alexander Kriz, Reno Van Durme, Benjamin Information Retrieval Computer Vision and Pattern Recognition Retrieval-augmented generation from videos requires systems to retrieve relevant audiovisual evidence from large corpora and synthesize it into coherent, attributed text. Current approaches struggle at both ends: retrieval methods fail on complex, multi-faceted queries that cannot be captured by a single embedding, while generation methods lack the high-level reasoning needed to synthesize across multiple videos and face memory constraints over long, multi-video contexts. We present MARQUIS: a three-stage pipeline that addresses these limitations through (1) query expansion, fusion, and reranking, (2) calibrated structured evidence extraction, and (3) article generation from extracted evidence, optionally controlled by an RLM. On the MAGMaR2026 shared task, we improve retrieval performance from 0.195 to 0.759 (nDCG@10). For article generation, ITER-QA-BASE improves average human score from 3.09 to 3.83 over the CAG baseline, while MARQUIS-RLM achieves a human score of 3.30 and the strongest citation recall among non-QA systems. |
| title | MARQUIS: A Three-Stage Pipeline for Video Retrieval-Augmented Generation |
| topic | Information Retrieval Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.17640 |