# Vision-Language Models Learn Super Images for Efficient Partially Relevant Video Retrieval

arXiv: https://arxiv.org/abs/2312.00414 · HTML: https://mlsanity.com/paper/2312.00414/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Video understanding with MLLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 50 of its 71 references that are in the map.
- First posted on arXiv 1 Dec 2023.

## Placed in

- Multimodal LLMs / vision-language models / [Video understanding with MLLMs](https://mlsanity.com/mllm/video-understanding-with-mllms/)

## Builds on (most cited first, 30 of 50)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Swin Transformer: Hierarchical Vision Transformer using Shifted Windows](https://mlsanity.com/paper/2103.14030/) (2021)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Scaling Laws for Neural Language Models](https://mlsanity.com/paper/2001.08361/) (2020)
- [Neural Machine Translation by Jointly Learning to Align and Translate](https://mlsanity.com/paper/1409.0473/) (2014)
- [Parameter-Efficient Transfer Learning for NLP](https://mlsanity.com/paper/1902.00751/) (2019)
- [Quo Vadis, Action Recognition? A New Model and the Kinetics Dataset](https://mlsanity.com/paper/1705.07750/) (2017)
- [CIDEr: Consensus-based Image Description Evaluation](https://mlsanity.com/paper/1411.5726/) (2014)
- [CoCa: Contrastive Captioners are Image-Text Foundation Models](https://mlsanity.com/paper/2205.01917/) (2022)
- [Frozen in Time: A Joint Video and Image Encoder for End-to-End Retrieval](https://mlsanity.com/paper/2104.00650/) (2021)
- [CLIP-Adapter: Better Vision-Language Models with Feature Adapters](https://mlsanity.com/paper/2110.04544/) (2021)
- [Dense-Captioning Events in Videos](https://mlsanity.com/paper/1705.00754/) (2017)
- [EVA: Exploring the Limits of Masked Visual Representation Learning at Scale](https://mlsanity.com/paper/2211.07636/) (2022)
- [CLIP4Clip: An Empirical Study of CLIP for End to End Video Clip Retrieval](https://mlsanity.com/paper/2104.08860/) (2021)
- [Towards Automatic Learning of Procedures from Web Instructional Videos](https://mlsanity.com/paper/1703.09788/) (2017)
- [VideoCLIP: Contrastive Pre-training for Zero-shot Video-Text Understanding](https://mlsanity.com/paper/2109.14084/) (2021)
- [TALL: Temporal Activity Localization via Language Query](https://mlsanity.com/paper/1705.02101/) (2017)
- [End-to-End Learning of Visual Representations from Uncurated Instructional Videos](https://mlsanity.com/paper/1912.06430/) (2019)
- [Less is More: ClipBERT for Video-and-Language Learning via Sparse Sampling](https://mlsanity.com/paper/2102.06183/) (2021)
- [VATEX: A Large-Scale, High-Quality Multilingual Dataset for Video-and-Language Research](https://mlsanity.com/paper/1904.03493/) (2019)
- [TVR: A Large-Scale Dataset for Video-Subtitle Moment Retrieval](https://mlsanity.com/paper/2001.09099/) (2020)
- [A Dataset for Movie Description](https://mlsanity.com/paper/1501.02530/) (2015)
- [Frozen CLIP Models are Efficient Video Learners](https://mlsanity.com/paper/2208.03550/) (2022)
- [Fine-grained Video-Text Retrieval with Hierarchical Graph Reasoning](https://mlsanity.com/paper/2003.00392/) (2020)
- [All in One: Exploring Unified Video-Language Pre-training](https://mlsanity.com/paper/2203.07303/) (2022)
- [SwinBERT: End-to-End Transformers with Sparse Attention for Video Captioning](https://mlsanity.com/paper/2111.13196/) (2021)

## Built on by (most cited first, 5 of 5)

- [Video Panels for Long Video Understanding](https://mlsanity.com/paper/2509.23724/) (2025)
- [Mitigating Semantic Collapse in Partially Relevant Video Retrieval](https://mlsanity.com/paper/2510.27432/) (2025)
- [Q2E: Query-to-Event Decomposition for Zero-Shot Multilingual Text-to-Video Retrieval](https://mlsanity.com/paper/2506.10202/) (2025)
- [Dual Learning with Dynamic Knowledge Distillation and Soft Alignment for Partially Relevant Video Retrieval](https://mlsanity.com/paper/2510.12283/) (2025)
- [ProPy: Building Interactive Prompt Pyramids upon CLIP for Partially Relevant Video Retrieval](https://mlsanity.com/paper/2508.19024/) (2025)

## Abstract (arXiv)

In this paper, we propose an efficient and high-performance method for partially relevant video retrieval, which aims to retrieve long videos that contain at least one moment relevant to the input text query. The challenge lies in encoding dense frames using visual backbones. This requires models to handle the increased frames, resulting in significant computation costs for long videos. To mitigate the costs, previous studies use lightweight visual backbones, yielding sub-optimal retrieval performance due to their limited capabilities. However, it is undesirable to simply replace the backbones with high-performance large vision-and-language models (VLMs) due to their low efficiency. To address this dilemma, instead of dense frames, we focus on super images, which are created by rearranging the video frames in an N × N grid layout. This reduces the number of visual encodings to frac1N2 and mitigates the low efficiency of large VLMs. Based on this idea, we make two contributions. First, we explore whether VLMs generalize to super images in a zero-shot setting. To this end, we propose a method called query-attentive super image retrieval (QASIR), which attends to partial moments relevant to the input query. The zero-shot QASIR yields two discoveries: (1) it enables VLMs to generalize to super images and (2) the grid size N, image resolution, and VLM size are key trade-off parameters between performance and computation costs. Second, we introduce fine-tuning and hybrid QASIR that combines high- and low-efficiency models to strike a balance between performance and computation costs. This reveals two findings: (1) the fine-tuning QASIR enhances VLMs to learn super images effectively, and (2) the hybrid QASIR minimizes the performance drop of large VLMs while reducing the computation costs.