arXiv:2606. 00987v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) have shown strong visual understanding and language-guided grounding abilities, yet their capacity for multi-temporal visual reasoning remains underexplored.
arXiv:2608. 01856v2 Announce Type: replace Abstract: Bi-temporal remote-sensing disaster change captioning often needs to identify sparse and spatially localized changes across large pre- and post-event scenes and then translate them into coherent, factual descriptions.
arXiv:2606. 28724v1 Announce Type: cross Abstract: Understanding and localizing subtle changes between paired images is critical for tasks such as surveillance and image editing.
The paper proposes a weakly supervised remote sensing change detection method that uses change captions as the sole supervision signal, eliminating the need for pixel‑level change masks. It introduces a caption‑driven generation pipeline to create bi‑temporal image pairs with controlled changes and a Semantic‑Appearance Agreement Framework (SAAF) that fuses caption‑grounded semantic responses with RGB differences for accurate change localization. Experiments on the Flair‑RSGen and WHU‑CDC datasets demonstrate that SAAF outperforms existing limited‑supervision baselines in macro‑averaged IoU and F1 metrics.
arXiv:2609.10356v1 Announce Type: new Abstract: Long-term change understanding from images of the same place revisited over time is a challenging task with applications in map maintenance and urban i...
The paper introduces MGRL-RSCC, a multi‑granularity reward reinforcement learning framework for Remote Sensing Change Captioning (RSCC). It uses a CNN with hierarchical self‑attention to extract visual features, a Transformer decoder for visual‑to‑linguistic translation, and a dual‑decoding strategy combined with token‑level supervised learning and self‑critical reinforcement learning. Three reward functions—linguistic fluency, change state consistency, and structural‑semantic relevance—are employed to improve caption quality and reduce exposure bias and conservative generation.
arXiv:2604. 20623v2 Announce Type: replace-cross Abstract: Traditional change detection identifies where changes occur, but does not explain what changed in natural language.
Long-term change understanding from images of the same place revisited over time is a challenging task with applications in map maintenance and urban infrastructure monitoring. Prior work addresses it...
CompoSET is a new vision‑language benchmark that tests compositionality by presenting each image‑caption pair with a foil that differs by only a single controlled edit—such as color, count, spatial relation, or object state—while keeping the overall scene largely unchanged. The dataset contains 1,776 instances across 80 scenes, 16 edit types in four semantic categories, and captions at three verbosity levels, with an average of only 8% pixel change between base and foil images. Evaluation of 22 dual‑encoder VLMs shows that all models perform at or below chance on CompoSET, contrasting with their strong performance on prior benchmarks that conflate global image differences with compositional reasoning.
The paper introduces FlipDir, a training‑free inference‑time technique that mitigates answer flips in vision‑language models by steering hidden states along a low‑rank subspace derived from contrastive image pairs. It employs a margin‑based gate to attenuate steering only during uncertain decoding steps, thereby restoring original predictions while keeping stable ones unchanged. The authors also present VisFlip, a benchmark framework that evaluates models across nine dataset‑variation combinations in scientific reasoning, robot‑scene understanding, and medical VQA, showing that FlipDir consistently outperforms existing methods on recovery and preservation metrics.
arXiv:2509.23928v3 Announce Type: replace-cross Abstract: Speculative decoding has proven effective for accelerating inference in Large Language Models (LLMs), yet its extension to Vision-Language Mo...
arXiv:2605.17949v2 Announce Type: replace Abstract: Remote sensing vision-language models (RS-VLMs) commonly employ a pretrained vision encoder and a projection module to map image features into the...