# MARQUIS: A Three-Stage Pipeline for Video Retrieval-Augmented Generation

Source: [MARQUIS: A Three-Stage Pipeline for Video Retrieval-Augmented Generation](<https://www.sota2.com/research/paper/6a0beee9dbc836b87d82c83a>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-20T05:00:13.583Z

Authors: Debashish Chakraborty, Dengjia Zhang, Jialiang Jin, Hanting Liu, Katherine Guerrerio, Hanxiang Qin, Tyler Skow, Alexander Martin, Reno Kriz, Benjamin Van Durme

Published: 2026-05-17T20:19:04.000Z

## Abstract

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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Article Generation on MAGMaR oracle (leaderboard snapshot)](<https://www.sota2.com/research/sota/article-generation-on-magmar-oracle-leaderboard-snapshot.md>): Article Generation; MAGMaR oracle (leaderboard snapshot); Human Preference Score: 3.833.

- [Video Retrieval on MAGMaR 2026](<https://www.sota2.com/research/sota/video-retrieval-on-magmar-2026.md>): Video Retrieval; MAGMaR 2026; nDCG@10: not recorded.

- [Information Retrieval on MAGMaR (final)](<https://www.sota2.com/research/sota/information-retrieval-on-magmar-final.md>): Information Retrieval; MAGMaR (final); Average Score: not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2605.17640>)

[PDF](<https://arxiv.org/pdf/2605.17640.pdf>)
