# Learning Question-Aware Keyframe Selection with Synthetic Supervision for Video Question Answering

arXiv: https://arxiv.org/abs/2603.14953 · HTML: https://mlsanity.com/paper/2603.14953/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Video understanding with MLLMs, one of 16 directions in Multimodal LLMs / vision-language models.
- Builds on 19 of its 20 references that are in the map.
- First posted on arXiv 16 Mar 2026.

## Placed in

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

## Builds on (most cited first, 19 of 19)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Visual Instruction Tuning](https://mlsanity.com/paper/2304.08485/) (2023)
- [Qwen2.5-VL Technical Report](https://mlsanity.com/paper/2502.13923/) (2025)
- [InstructBLIP: Towards General-purpose Vision-Language Models with Instruction Tuning](https://mlsanity.com/paper/2305.06500/) (2023)
- [Video-LLaMA: An Instruction-tuned Audio-Visual Language Model for Video Understanding](https://mlsanity.com/paper/2306.02858/) (2023)
- [EVA-CLIP: Improved Training Techniques for CLIP at Scale](https://mlsanity.com/paper/2303.15389/) (2023)
- [NExT-QA:Next Phase of Question-Answering to Explaining Temporal Actions](https://mlsanity.com/paper/2105.08276/) (2021)
- [CLEVRER: CoLlision Events for Video REpresentation and Reasoning](https://mlsanity.com/paper/1910.01442/) (2019)
- [Zero-Shot Video Question Answering via Frozen Bidirectional Language Models](https://mlsanity.com/paper/2206.08155/) (2022)
- [Adaptive Keyframe Sampling for Long Video Understanding](https://mlsanity.com/paper/2502.21271/) (2025)
- [AGQA: A Benchmark for Compositional Spatio-Temporal Reasoning](https://mlsanity.com/paper/2103.16002/) (2021)
- [Frame-Voyager: Learning to Query Frames for Video Large Language Models](https://mlsanity.com/paper/2410.03226/) (2024)
- [Video as Conditional Graph Hierarchy for Multi-Granular Question Answering](https://mlsanity.com/paper/2112.06197/) (2021)
- [Invariant Grounding for Video Question Answering](https://mlsanity.com/paper/2206.02349/) (2022)
- [Bridge to Answer: Structure-aware Graph Interaction Network for Video Question Answering](https://mlsanity.com/paper/2104.14085/) (2021)
- [KeyVideoLLM: Towards Large-scale Video Keyframe Selection](https://mlsanity.com/paper/2407.03104/) (2024)
- [Discovering Spatio-Temporal Rationales for Video Question Answering](https://mlsanity.com/paper/2307.12058/) (2023)
- [End-to-End Video Question Answering with Frame Scoring Mechanisms and Adaptive Sampling](https://mlsanity.com/paper/2407.15047/) (2024)
- [Weakly Supervised Gaussian Contrastive Grounding with Large Multimodal Models for Video Question Answering](https://mlsanity.com/paper/2401.10711/) (2024)

## Abstract (arXiv)

Large multimodal models (LMMs) have recently demonstrated remarkable performance in video question answering (VideoQA), yet reasoning over video remains challenging due to high inference cost and diluted information. Keyframe selection offers efficiency and sharper reasoning but suffers from sparse supervision and redundant frame choices when relying only on image-text similarity. We present a question-aware keyframe selection framework with two components: pseudo keyframe labels derived from LMMs that provide informative supervision and a coverage regularization that promotes diverse, complementary evidence across time. Experiments on NExT-QA show that our method significantly improves accuracy, especially for temporal and causal question types, establishing keyframe selection as an effective and learnable module for VideoQA.