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

TL;DR

MARQUIS introduces a three-stage pipeline to enhance video retrieval and generation, improving retrieval accuracy and article quality by addressing current limitations in multi-video reasoning and memory constraints.

Contribution

It proposes a novel three-stage pipeline for video retrieval-augmented generation, combining query expansion, evidence extraction, and controlled article synthesis.

Findings

Retrieval performance improved from 0.195 to 0.759 (nDCG@10).

Human scores increased from 3.09 to 3.83 with ITER-QA-BASE.

MARQUIS-RLM achieved a human score of 3.30 and strong citation recall.

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.