# Revisiting Change VQA in Remote Sensing with Structured and Native Multimodal Qwen Models

arXiv: https://arxiv.org/abs/2604.18429 · HTML: https://mlsanity.com/paper/2604.18429/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 11 of its 16 references that are in the map.
- First posted on arXiv 20 Apr 2026.

## Builds on (most cited first, 11 of 11)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Qwen3-VL Technical Report](https://mlsanity.com/paper/2511.21631/) (2025)
- [LlamaFactory: Unified Efficient Fine-Tuning of 100+ Language Models](https://mlsanity.com/paper/2403.13372/) (2024)
- [RSVQA: Visual Question Answering for Remote Sensing Data](https://mlsanity.com/paper/2003.07333/) (2020)
- [EarthVQA: Towards Queryable Earth via Relational Reasoning-Based Remote Sensing Visual Question Answering](https://mlsanity.com/paper/2312.12222/) (2023)
- [Change Detection Meets Visual Question Answering](https://mlsanity.com/paper/2112.06343/) (2021)
- [HRVQA: A Visual Question Answering Benchmark for High-Resolution Aerial Images](https://mlsanity.com/paper/2301.09460/) (2023)
- [Vision-Language Modeling Meets Remote Sensing: Models, Datasets and Perspectives](https://mlsanity.com/paper/2505.14361/) (2025)
- [Show Me What and Where has Changed? Question Answering and Grounding for Remote Sensing Change Detection](https://mlsanity.com/paper/2410.23828/) (2024)
- [DeltaVLM: Interactive Remote Sensing Image Change Analysis via Instruction-guided Difference Perception](https://mlsanity.com/paper/2507.22346/) (2025)
- [Text-conditioned State Space Model For Domain-generalized Change Detection Visual Question Answering](https://mlsanity.com/paper/2508.08974/) (2025)

## Built on by (most cited first, 1 of 1)

- [Selective Tool Use for Agentic Change Visual Question Answering in Remote Sensing](https://mlsanity.com/paper/2609.14523/) (2026)

## Abstract (arXiv)

Change visual question answering (Change VQA) addresses the problem of answering natural-language questions about semantic changes between bi-temporal remote sensing (RS) images. Although vision-language models (VLMs) have recently been studied for temporal RS image understanding, Change VQA remains underexplored in the context of modern multimodal models. In this letter, we revisit the CDVQA benchmark using recent Qwen models under a unified low-rank adaptation (LoRA) setting. We compare Qwen3-VL, which follows a structured vision-language pipeline with multi-depth visual conditioning and a full-attention decoder, with Qwen3.5, a native multimodal model that combines a single-stage alignment with a hybrid decoder backbone. Experimental results on the official CDVQA test splits show that recent VLMs improve over earlier specialized baselines. They further show that performance does not scale monotonically with model size, and that native multimodal models are more effective than structured vision-language pipelines for this task. These findings indicate that tightly integrated multimodal backbones contribute more to performance than scale or explicit multi-depth visual conditioning for language-driven semantic change reasoning in RS imagery.