Pith. sign in

REVIEW 3 major objections 4 minor 4 cited by

RSUniVLM: A Unified Vision Language Model for Remote Sensing via Granularity-oriented Mixture of Experts

T0 review · 3 major / 4 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read RSUniVLM claims to be the first remote-sensing vision-language model to handle image-, region-, and pixel-level tasks in one 1B-parameter system, with state-of-the-art grounding and a large zero-shot segmentation gain.

desk verdict A genuinely useful unification of RS VLM granularities with a clean MoE design, but the headline numbers need split disclosures before they mean anything. read the letter →

arxiv 2412.05679 v2 pith:Z45EIR46 submitted 2024-12-07 cs.CV

Xu Liu , Zhouhui Lian This is my paper · ORCID
classification cs.CV
keywords remotesensingvisionlanguagemodelmixtureofexpertsvisualgroundingsemanticsegmentationchangedetectioncaptioninginstructiontuning
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper sets out to show that a single remote-sensing vision-language model can cover every granularity of visual understanding, including whole-image captioning and question answering, region-level object localization, pixel-level segmentation, and multi-image change analysis, all without task-specific heads or a large parameter count. To do this, it builds RSUniVLM, a roughly 1B-parameter model whose language backbone is split into three experts by a Granularity-oriented Mixture of Experts router, and whose outputs are always text: bounding boxes are written as coordinate strings and masks as semantic descriptors. On the reported benchmarks, the model reaches 72.47 acc@0.5 on DIOR-RSVG and 69.31 on VRSBench-Ref for visual grounding, and 56.49 average mIoU on zero-shot semantic segmentation, a 35% relative gain over the strongest compared open-vocabulary segmenter. The paper's claim matters because it suggests that a compact, end-to-end unified model can replace an array of specialized remote-sensing systems, and that segmentation can be folded into ordinary text generation.

What carries the argument

The load-bearing mechanism is the Granularity-oriented Mixture of Experts (G-MoE), which duplicates the language model's feed-forward layers into three experts: one for image-level semantics, one for localized region understanding, and one for pixel-level detail, and routes each instruction to the matching expert with a training-free router. Complementing this is a unified text-only output scheme: bounding boxes are emitted as integer coordinate strings in the range [0, 100], and segmentation masks are serialized as 24x24 rows of semantic descriptors using the run-length representation from Text4Seg. Together, these choices make every task, including mask generation, an ordinary language-model next-token prediction, so the model can be trained end-to-end without extra heads.

What would settle it

Audit the released training mixtures against the evaluation splits of VRSBench-Ref and DIOR-RSVG: any overlap of images or annotated regions between the 141k/30k training samples and the test sets would invalidate the headline grounding accuracies of 69.31 and 72.47.

Watch

Extended reading notes

Core claim

The central claim is that a unified text-only representation plus a granularity-decoupled mixture of experts lets one vision-language model master tasks at three levels of visual detail at once. RSUniVLM is presented as the first remote-sensing VLM that jointly supports scene classification, visual question answering, captioning, visual grounding, referring expression generation, semantic segmentation, change captioning, and change detection, all in an end-to-end autoregressive framework. The paper reports that the model outperforms prior remote-sensing VLMs on visual grounding by a wide margin, matches specialized change-captioning systems on LEVIR-MCI, and improves zero-shot semantic segmentation over CLIP-based open-vocabulary baselines by roughly 35% relative on average, while keeping about 1 billion total parameters and only 313 million trainable in the G-MoE stage.

Load-bearing premise

The results assume that the public datasets used for training are cleanly separated from those used for scoring, and that the zero-shot segmentation and change-detection evaluations measure transfer to genuinely unseen data rather than overlap with training material.

Editorial extensions

If this is right

  • If the claims hold, remote-sensing change detection and semantic segmentation no longer require dedicated decoders; both can be produced as text from the same autoregressive model.
  • A 1B-parameter unified model that beats 7B grounding specialists suggests that parameter count is not the main driver of fine-grained RS perception, and that granularity-decoupled routing is a viable alternative to scaling.
  • Zero-shot semantic segmentation at 56.49 average mIoU would reposition open-vocabulary RS segmentation as a language-model capability rather than a CLIP-specific one.
  • The reported change-captioning parity with specialized systems implies that multi-image reasoning and single-image reasoning can share one weight set without catastrophic interference.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference: A direct test the paper leaves implicit is ablating the G-MoE router by forcing all prompts through a single expert; if accuracy stays flat, the granularity decoupling is not what drives the reported gains.
  • Editorial inference: Because G-MoE splits by granularity rather than by task, the same architecture could be carried to other dense-perception domains, such as medical imaging or autonomous driving, where whole-scene, object, and pixel understanding are currently separate models.
  • Editorial inference: The paper's stated weak multi-turn conversation suggests a follow-up test: mixing multi-turn RS instruction data into stage-2 training should improve conversation without degrading grounding; if grounding degrades, the granularity split is absorbing part of the trade-off.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper introduces RSUniVLM, a ~1B-parameter remote-sensing vision-language model that unifies image-level (classification, VQA, captioning), region-level (visual grounding, referring expression generation), pixel-level (semantic segmentation), and multi-image (change captioning/change detection) tasks in a single end-to-end architecture. The key technical contribution is a Granularity-oriented Mixture of Experts (G-MoE) with three task-type experts and a training-free router, together with a two-stage training scheme over a newly assembled instruction-following dataset. The paper reports state-of-the-art or competitive results across 13 datasets, with the strongest claims being visual grounding on DIOR-RSVG (72.47 acc@0.5) and VRSBench-Ref (69.31 acc@0.5), change captioning on LEVIR-MCI (CIDEr 139.80), and zero-shot semantic segmentation (56.49 mIoU average).

Significance. If the reported numbers are valid, RSUniVLM is a valuable proof-of-concept that a relatively small (1B) RS VLM can cover many visual granularities, and the G-MoE routing idea is simple and worth exploring. The paper is strong in reproducibility infrastructure: it provides detailed dataset statistics, training hyperparameters, prompt templates, and qualitative examples. The breadth of tasks unified in one model is impressive. However, the empirical foundation is currently compromised by the absence of train/test split disclosure for benchmarks that also appear in the training data, and by the misleading use of the term 'zero-shot' for tasks on which the model was directly supervised. These issues must be resolved before the state-of-the-art claims can be trusted.

major comments (3)
  1. [Section 4.1, 4.2; Supplementary Tables 10 and 12] The visual-grounding and change-captioning evaluations in Tables 4 and 5 use DIOR-RSVG, VRSBench-Ref, and LEVIR-MCI, all of which are listed in the training data (stage 1 and stage 2) without any specification of train/test splits. In particular, VRSBench-Ref is a component of VRSBench, and LEVIR-MCI CC and CD are both in the stage-2 data. Because no split is described, the reported acc@0.5 of 72.47 on DIOR-RSVG, 69.31 on VRSBench-Ref, and CIDEr 139.80 on LEVIR-MCI may reflect memorization rather than generalization. Please specify exactly which images were used for training and for evaluation (e.g., official split names and counts), and if any evaluation image was seen during training, re-run the evaluations on a strictly disjoint subset.
  2. [Section 4.2, Table 7] The 'zero-shot' semantic segmentation comparison in Table 7 is not zero-shot with respect to the task: RSUniVLM was trained on multiple semantic-segmentation datasets (Potsdam, UA Vid, LoveDA, OpenEarthMap, landcover, COCO stuff) and is prompted to produce masks for classes such as building, road, and vegetation. The baselines (MaskCLIP, SCLIP, GEM, ClearCLIP, SegEarth-OV) are training-free and have never seen segmentation supervision. The 35% mIoU gain over SegEarth-OV therefore largely reflects task-specific supervised training, not emergent zero-shot ability. Please re-frame this as cross-dataset or unseen-dataset transfer and compare with models that receive the same segmentation supervision, or remove the zero-shot claim.
  3. [Section 4.2, Table 6 and Supplementary Table 12] The change-detection evaluation on WHU-CD is also labeled 'zero-shot,' but LEVIR-MCI change-detection data are part of the stage-2 training set (Supplementary Table 12). The comparison to specialized models trained with 5% labeled samples is informative as cross-dataset transfer, but it should not be described as zero-shot, since the model has been fine-tuned on the same task type. Please adjust the terminology and clarify the relation between the training and evaluation distributions.
minor comments (4)
  1. [Abstract and Table 4] The abstract states that RSUniVLM 'far exceeds' GeoChat's 19.1% on VRSBench-Ref, but Table 4 lists GeoChat's acc@0.7 as 19.1 and acc@0.5 as 39.6; comparing 69.31 (acc@0.5) against 19.1 (acc@0.7) mixes thresholds and should be corrected.
  2. [Section 4.1 and Supplementary Tables 10-12] The paper says the stage-1 dataset contains 1.2M instructions, but the numbers in Supplementary Tables 10 and 11 sum to approximately 1.53M (985k + 547k); please reconcile this discrepancy or state the sampling procedure.
  3. [Table 5 caption] The caption begins 'Change Captioning performance comparison ... we employ BLEU...' with a lowercase 'we' after the period; please fix the capitalization.
  4. [Section 3.2 and Abstract] The G-MoE router is described as 'training-free' and assigns prompts to experts based on task identifiers; this is a deterministic task-conditional routing rather than a learned mixture-of-experts gating, and the paper should clarify this in the method section. Also, the phrase 'without increasing model size' in the abstract is inaccurate because duplicating the FFN layers three times increases total parameters; the claim should be restricted to inference-time computation or runtime parameters.

Circularity Check

2 steps flagged · score 6.0 of 10

As written, the benchmark results that support 'state-of-the-art' are computed on datasets listed in the training mix with no disclosed split, making the headline numbers fit scores rather than predictions.

  1. fitted input called prediction [Section 4.1 (Dataset Details), Supplementary Tables 10 and 12, Section 4.2 (Visual Grounding), Table 4]
    "For region-level tasks, we utilize two prominent open-access visual grounding datasets, DIOR-RSVG [86] and OPT-RSVG [39]. ... We use all the aforementioned data for the first stage of model training. ... Visual Grounding. We use DIOR-RSVG [86] and VRSBench-Ref [40] for the evaluation of visual grounding."

    The same named benchmarks used for evaluation, DIOR-RSVG and VRSBench-Ref, are listed in the stage-1 and stage-2 training tables (Supplementary Tables 10 and 12), and no train/val/test split is described anywhere in the paper. As written, therefore, the model's parameters are optimized on the very samples whose grounding accuracy Table 4 reports; the 72.47 and 69.31 acc@0.5 numbers are measures of training-set fit, not predictions on held-out data. The ablation in Table 8 averages the same two benchmarks, so the claimed G-MoE advantage is likewise evaluated on in-distribution training data.

  2. fitted input called prediction [Section 4.1, Section 4.2 (Change Captioning), Table 5, Supplementary Tables 10 and 12]
    "Besides, we include LEVIR-MCI [49] to improve multi-image comprehension. ... LEVIR-MCI [49] CC & CD 40k ... Image-level ... LEVIR-MCI [49] CC ... Change Captioning performance comparison on LEVIR-MCI [49]."

    LEVIR-MCI appears in both the training mixture (Supplementary Table 10: 40k CC&CD samples; Supplementary Table 12: image-level CC and pixel-level CD) and as the evaluation set for Table 5's change-captioning numbers. With no split specified, the CIDEr 139.80 is computed on samples the model was trained on, so the claim of being comparable to task-specific models is a fit report rather than an independent prediction.

full rationale

RSUniVLM is an empirical systems paper, not a derivation; most architectural choices (SigLIP encoder, Qwen2 LLM, Text4Seg descriptors, G-MoE) are stated as adopted components and are not circular. No load-bearing self-citation appears: the authors do not rely on their own prior theorems. The circularity is concentrated in the benchmark protocol: the headline grounding, change-captioning, and ablation numbers are computed on datasets that the paper's own data tables list as training data, and the paper never discloses a train/test split. As written, those numbers reduce to memorization of training inputs. The zero-shot segmentation and WHU-CD change-detection results use different datasets from the training set, so they are not circular by construction, though their 'zero-shot' label is a weaker claim than the training-free baselines; that is a correctness/external-validity concern, not circularity. Overall score 6 reflects partial circularity in the central quantitative claims while the unified-model capability claim retains independent content.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The central claims rest on standard supervised-learning assumptions and a few hand-picked design choices (expert count, mask grid, task-token routing, stage-2 data mix, LoRA rank). No new physical or conceptual entities are introduced. The main unstated load is the assumption that the public benchmarks used for training and evaluation are properly split and that zero-shot evaluations are meaningful despite same-task training.

free parameters (6)
  • Number of granularity experts = 3
    Chosen to match the three granularity levels (image, region, pixel); not optimized on validation data.
  • Semantic descriptor grid size = 24x24
    Masks are encoded as 24x24 semantic descriptors following Text4Seg; this resolution limits segmentation precision and is a design choice that affects all pixel-level results.
  • Task router assignment = Hard mapping from task token to expert
    The training-free router assigns each prompt to an expert based on the task token; this design choice determines which tasks share parameters and is not learned or validated.
  • Stage-2 dataset resampling = Per-granularity subsets listed in Table 12
    The composition of stage-2 data is selected by hand and affects G-MoE fine-tuning; the sampling ratios and exclusion criteria are not justified.
  • Baseline MoE expert count = 8
    Number of experts for the basic MoE baseline in the ablation; the choice is not justified and may disadvantage the baseline.
  • LoRA rank and scale (ablation) = r=128, alpha=256
    Hyperparameters for the LoRA baseline in the ablation; chosen by hand and not swept.
assumptions (4)
  • domain assumption Supervised instruction tuning on mixed public datasets transfers to held-out RS benchmarks
    The entire evaluation assumes that training on the collected 1.2M instructions improves generalization rather than overfitting; this is standard in the field but not proven. Invoked throughout Section 4.
  • domain assumption Text4Seg semantic descriptors provide a valid mask representation for training and evaluation
    Pixel-level tasks rely on 24x24 descriptor sequences; the fidelity of this representation for precise segmentation is assumed. Invoked in Section 3.1 and Section 4.1.
  • domain assumption Task tokens are sufficient to route inputs to the right expert
    G-MoE's training-free router uses task tokens; it assumes that task identity is always available at inference and that prompts are not mixed-granularity. Invoked in Section 3.2.
  • domain assumption Public datasets used for training have labels consistent with evaluation conventions
    The model inherits label noise and class definitions from DIOR-RSVG, VRSBench, LEVIR-MCI, Potsdam, and others. The paper does not audit label quality. Invoked in Section 4.1.

how reviews work

0 comments
Cite this review

Pith. "Pith review of RSUniVLM: A Unified Vision Language Model for Remote Sensing via Granularity-oriented Mixture of Experts." pith.science (2026). https://pith.science/paper/Z45EIR46

@misc{pith2026241205679,
  author       = {Pith},
  title        = {Pith review of: RSUniVLM: A Unified Vision Language Model for Remote Sensing via Granularity-oriented Mixture of Experts},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Z45EIR46}},
  note         = {Machine review of arXiv:2412.05679}
}
read the original abstract

Remote Sensing Vision-Language Models (RS VLMs) have made much progress in the tasks of remote sensing (RS) image comprehension. While performing well in multi-modal reasoning and multi-turn conversations, the existing models lack pixel-level understanding and struggle with multi-image inputs. In this work, we propose RSUniVLM, a unified, end-to-end RS VLM designed for comprehensive vision understanding across multiple granularity, including image-level, region-level, and pixel-level tasks. RSUniVLM also performs effectively in multi-image analysis, with instances of change detection and change captioning. To enhance the model's ability to capture visual information at different levels without increasing model size, we design a novel architecture called Granularity-oriented Mixture of Experts to constraint the model to about 1 billion parameters. We also construct a large-scale RS instruction-following dataset based on a variety of existing datasets in both RS and general domain, encompassing various tasks such as object localization, visual question answering, and semantic segmentation. Substantial experiments have been conducted to validate the superiority of the proposed RSUniVLM up to state-of-the-art across various RS tasks. Code and model will be available at \href{https://github.com/xuliu-cyber/RSUniVLM}{here}.

Figures

Figures reproduced from arXiv: 2412.05679 by the authors.

Figure 1
Figure 1. RSUniVLM is a unified remote sensing VLM with versatile capabilities across three levels of visual understanding: a) Image cap [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. An overview of RSUniVLM. We adopt the classic LLaVA-based architecture, which consists of an image encoder, an MLP [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. Qualitative Results of RSUniVLM across a variety of tasks, demonstrating the ability of RSUniVLM to handle multi-level visual [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (3 more)
Figure 4
Figure 4. Figure 4: Qualitative Results of visual grounding. [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]
Figure 5
Figure 5. Figure 5: Qualitative Results of semantic segmentation. [PITH_FULL_IMAGE:figures/full_fig_p015_5.png]
Figure 6
Figure 6. Figure 6: Qualitative Results of change captioning and change detection. [PITH_FULL_IMAGE:figures/full_fig_p016_6.png]

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 4 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. SkyVLaM: Multimodal Large Language Model for UAV Video Understanding in Remote Sensing

    cs.CV 2026-07 conditional novelty 6.0 of 10

    SkyVLaM introduces a temporal basis perceiver and adaptive dense selection to improve language-conditioned video segmentation in UAV scenes, and contributes the SkyVid dataset.

  2. WeaveEarth: Structured Evidence Construction and Reasoning for Training-Free UHR Remote Sensing Understanding

    cs.CV 2026-07 conditional novelty 6.0 of 10

    A training-free evidence-selection and topology-preserving packaging pipeline improves frozen VLMs on ultra-high-resolution remote sensing VQA without multi-round search.

  3. Remote Sensing Large Vision-Language Model: Semantic-augmented Multi-level Alignment and Semantic-aware Expert Modeling

    cs.CV 2025-06 conditional novelty 5.0 of 10

    A remote sensing LVLM that augments visual features with retrieved captions and routes them through level-specific experts improves performance on several RS vision-language benchmarks.

  4. Domain Specific Benchmarks for Evaluating Multimodal Large Language Models

    cs.LG 2025-06 conditional novelty 3.0 of 10

    A review paper that organizes domain-specific MLLM benchmarks into an eight-discipline taxonomy, with summary tables and performance highlights.

Reference graph

Works this paper leans on

102 extracted references · 36 canonical work pages · cited by 4 Pith papers

  1. [1]

    Flamingo: a visual language model for few-shot learning

    Jean-Baptiste Alayrac, Jeff Donahue, Pauline Luc, Antoine Miech, Iain Barr, Yana Hasson, Karel Lenc, Arthur Men- sch, Katherine Millican, Malcolm Reynolds, et al. Flamingo: a visual language model for few-shot learning. Advances in neural information processing systems, 35:23716–23736,

  2. [2]

    Qwen-vl: A frontier large vision-language model with versatile abilities

    Jinze Bai, Shuai Bai, Shusheng Yang, Shijie Wang, Sinan Tan, Peng Wang, Junyang Lin, Chang Zhou, and Jingren Zhou. Qwen-vl: A frontier large vision-language model with versatile abilities. arXiv preprint arXiv:2308.12966, 2023. 2, 3, 5

    arXiv 2023

  3. [3]

    Wele Gedara Chaminda Bandara and Vishal M. Patel. Revis- iting consistency regularization for semi-supervised change detection in remote sensing images, 2022. 6

    2022

  4. [4]

    Vlmo: Unified vision-language pre-training with mixture-of-modality-experts

    Hangbo Bao, Wenhui Wang, Li Dong, Qiang Liu, Owais Khan Mohammed, Kriti Aggarwal, Subhojit Som, Songhao Piao, and Furu Wei. Vlmo: Unified vision-language pre-training with mixture-of-modality-experts. Advances in Neural Information Processing Systems , 35:32897–32912,

  5. [5]

    Land- cover

    Adrian Boguszewski, Dominik Batorski, Natalia Ziemba- Jankowska, Tomasz Dziedzic, and Anna Zambrzycka. Land- cover. ai: Dataset for automatic mapping of buildings, wood- lands, water and roads from aerial imagery. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pat- tern Recognition, pages 1102–1110, 2021. 6, 1, 5

    2021

  6. [6]

    Grounding everything: Emerging localiza- tion properties in vision-language transformers

    Walid Bousselham, Felix Petersen, Vittorio Ferrari, and Hilde Kuehne. Grounding everything: Emerging localiza- tion properties in vision-language transformers. In Proceed- ings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 3828–3837, 2024. 6

    2024

  7. [7]

    Coco- stuff: Thing and stuff classes in context

    Holger Caesar, Jasper Uijlings, and Vittorio Ferrari. Coco- stuff: Thing and stuff classes in context. In Proceedings of the IEEE conference on computer vision and pattern recog- nition, pages 1209–1218, 2018. 6, 5

    2018

  8. [8]

    Vdd: Varied drone dataset for semantic segmentation

    W Cai, K Jin, J Hou, C Guo, L Wu, and W Yang. Vdd: Varied drone dataset for semantic segmentation. arxiv 2023. arXiv preprint arXiv:2305.13608. 6, 8

    arXiv 2023

Show all 102 references
  1. [9]

    Changes to captions: An attentive network for remote sensing change captioning

    Shizhen Chang and Pedram Ghamisi. Changes to captions: An attentive network for remote sensing change captioning. IEEE Transactions on Image Processing, 2023. 6

    2023

  2. [10]

    Sharegpt4v: Improving large multi-modal models with better captions

    Lin Chen, Jinsong Li, Xiaoyi Dong, Pan Zhang, Conghui He, Jiaqi Wang, Feng Zhao, and Dahua Lin. Sharegpt4v: Improving large multi-modal models with better captions. arXiv preprint arXiv:2311.12793, 2023. 2, 6, 5

    2023 arXiv

  3. [11]

    Large-scale structure from motion with semantic con- straints of aerial images

    Yu Chen, Yao Wang, Peng Lu, Yisong Chen, and Guoping Wang. Large-scale structure from motion with semantic con- straints of aerial images. In Chinese Conference on Pattern Recognition and Computer Vision (PRCV) , pages 347–359. Springer, 2018. 6, 8

    2018

  4. [12]

    Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks

    Zhe Chen, Jiannan Wu, Wenhai Wang, Weijie Su, Guo Chen, Sen Xing, Muyan Zhong, Qinglong Zhang, Xizhou Zhu, Lewei Lu, Bin Li, Ping Luo, Tong Lu, Yu Qiao, and Jifeng Dai. Internvl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks. arXiv prepri...

    2023 arXiv

  5. [13]

    How far are we to gpt-4v? closing the gap to commercial multimodal models with open-source suites

    Zhe Chen, Weiyun Wang, Hao Tian, Shenglong Ye, Zhang- wei Gao, Erfei Cui, Wenwen Tong, Kongzhi Hu, Jiapeng Luo, Zheng Ma, et al. How far are we to gpt-4v? closing the gap to commercial multimodal models with open-source suites. arXiv preprint arXiv:2404.16821, 2024. 2, 3

    2024 arXiv

  6. [14]

    Remote sens- ing image scene classification: Benchmark and state of the art

    Gong Cheng, Junwei Han, and Xiaoqiang Lu. Remote sens- ing image scene classification: Benchmark and state of the art. Proceedings of the IEEE, 105(10):1865–1883, 2017. 6, 7, 1, 5

    2017

  7. [15]

    Functional map of the world

    Gordon Christie, Neil Fendley, James Wilson, and Ryan Mukherjee. Functional map of the world. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 6172–6180, 2018. 1, 5

    2018

  8. [16]

    Functional map of the world

    Gordon Christie, Neil Fendley, James Wilson, and Ryan Mukherjee. Functional map of the world. In CVPR, 2018. 6

    2018

  9. [17]

    Qwen-audio: Advancing universal audio understanding via unified large-scale audio-language models

    Yunfei Chu, Jin Xu, Xiaohuan Zhou, Qian Yang, Shil- iang Zhang, Zhijie Yan, Chang Zhou, and Jingren Zhou. Qwen-audio: Advancing universal audio understanding via unified large-scale audio-language models. arXiv preprint arXiv:2311.07919, 2023. 3

    2023 arXiv

  10. [18]

    Satellite image classification via two-layer sparse coding with biased image representa- tion

    Dengxin Dai and Wen Yang. Satellite image classification via two-layer sparse coding with biased image representa- tion. IEEE Geoscience and remote sensing letters, 8(1):173– 176, 2010. 7

    2010

  11. [19]

    Instructblip: Towards general- purpose vision-language models with instruction tuning,

    Wenliang Dai, Junnan Li, Dongxu Li, Anthony Meng Huat Tiong, Junqi Zhao, Weisheng Wang, Boyang Li, Pascale Fung, and Steven Hoi. Instructblip: Towards general- purpose vision-language models with instruction tuning,

  12. [20]

    Changechat: An interactive model for remote sensing change analy- sis via multimodal instruction tuning

    Pei Deng, Wenqian Zhou, and Hanlin Wu. Changechat: An interactive model for remote sensing change analy- sis via multimodal instruction tuning. arXiv preprint arXiv:2409.08582, 2024. 2, 3

    2024 arXiv

  13. [21]

    Qlora: Efficient finetuning of quantized llms

    Tim Dettmers, Artidoro Pagnoni, Ari Holtzman, and Luke Zettlemoyer. Qlora: Efficient finetuning of quantized llms. Advances in Neural Information Processing Systems , 36,

  14. [22]

    Vitron: A unified pixel-level vision llm for understanding, generating, segmenting, editing, 2024

    Hao Fei, Shengqiong Wu, Hanwang Zhang, Tat-Seng Chua, and Shuicheng Yan. Vitron: A unified pixel-level vision llm for understanding, generating, segmenting, editing, 2024. 2, 3

    2024

  15. [23]

    Rsteller: Scaling up visual language modeling in remote sensing with rich linguistic semantics from openly avail- able data and large language models

    Junyao Ge, Yang Zheng, Kaitai Guo, and Jimin Liang. Rsteller: Scaling up visual language modeling in remote sensing with rich linguistic semantics from openly avail- able data and large language models. arXiv preprint arXiv:2408.14744, 2024. 2, 6, 1

    2024 arXiv

  16. [24]

    Dbrx: Creating an llm from scratch using databricks

    Nikhil Gupta and Jason Yip. Dbrx: Creating an llm from scratch using databricks. In Databricks Data Intelligence Platform: Unlocking the GenAI Revolution, pages 311–330. Springer, 2024. 3

    2024

  17. [25]

    Retrieval augmented language model pre- training

    Kelvin Guu, Kenton Lee, Zora Tung, Panupong Pasupat, and Mingwei Chang. Retrieval augmented language model pre- training. In International conference on machine learning , pages 3929–3938. PMLR, 2020. 2, 3

    2020

  18. [26]

    Ai2d-rst: A multimodal corpus of 1000 primary school science dia- grams

    Tuomo Hiippala, Malihe Alikhani, Jonas Haverinen, Timo Kalliokoski, Evanfiya Logacheva, Serafina Orekhova, Aino Tuomainen, Matthew Stone, and John A Bateman. Ai2d-rst: A multimodal corpus of 1000 primary school science dia- grams. Language Resources and Evaluation , 55:661–688,

  19. [27]

    Lora: Low-rank adaptation of large language models

    Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen- Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. Lora: Low-rank adaptation of large language models. arXiv preprint arXiv:2106.09685, 2021. 2

    2021 arXiv

  20. [28]

    Rsgpt: A remote sensing vision language model and benchmark

    Yuan Hu, Jianlong Yuan, Congcong Wen, Xiaonan Lu, and Xiang Li. Rsgpt: A remote sensing vision language model and benchmark. arXiv preprint arXiv:2307.15266, 2023. 2, 3, 5

    2023 arXiv

  21. [29]

    Fully convolutional networks for multisource building extraction from an open aerial and satellite imagery data set

    Shunping Ji, Shiqing Wei, and Meng Lu. Fully convolutional networks for multisource building extraction from an open aerial and satellite imagery data set. IEEE Transactions on geoscience and remote sensing, 57(1):574–586, 2018. 6, 8

    2018

  22. [30]

    Mixtral of experts

    Albert Q Jiang, Alexandre Sablayrolles, Antoine Roux, Arthur Mensch, Blanche Savary, Chris Bamford, Deven- dra Singh Chaplot, Diego de las Casas, Emma Bou Hanna, Florian Bressand, et al. Mixtral of experts. arXiv preprint arXiv:2401.04088, 2024. 3

    2024 arXiv

  23. [31]

    Dvqa: Understanding data visualizations via ques- tion answering

    Kushal Kafle, Brian Price, Scott Cohen, and Christopher Kanan. Dvqa: Understanding data visualizations via ques- tion answering. In Proceedings of the IEEE conference on computer vision and pattern recognition, pages 5648–5656,

  24. [32]

    Geochat: Grounded large vision-language model for remote sensing

    Kartik Kuckreja, Muhammad Sohail Danish, Muzammal Naseer, Abhijit Das, Salman Khan, and Fahad Shahbaz Khan. Geochat: Grounded large vision-language model for remote sensing. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 27831– 2784...

    2024

  25. [33]

    Lisa: Reasoning segmentation via large language model

    Xin Lai, Zhuotao Tian, Yukang Chen, Yanwei Li, Yuhui Yuan, Shu Liu, and Jiaya Jia. Lisa: Reasoning segmentation via large language model. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 9579–9589, 2024. 2

    2024

  26. [34]

    Clearclip: Decom- posing clip representations for dense vision-language infer- ence

    Mengcheng Lan, Chaofeng Chen, Yiping Ke, Xinjiang Wang, Litong Feng, and Wayne Zhang. Clearclip: Decom- posing clip representations for dense vision-language infer- ence. arXiv preprint arXiv:2407.12442, 2024. 6, 8

    2024 arXiv

  27. [35]

    Text4seg: Reimagining image segmentation as text genera- tion

    Mengcheng Lan, Chaofeng Chen, Yue Zhou, Jiaxing Xu, Yiping Ke, Xinjiang Wang, Litong Feng, and Wayne Zhang. Text4seg: Reimagining image segmentation as text genera- tion. arXiv preprint arXiv:2410.09855, 2024. 2, 4, 6, 1

    2024 arXiv

  28. [36]

    Llava-onevision: Easy visual task transfer

    Bo Li, Yuanhan Zhang, Dong Guo, Renrui Zhang, Feng Li, Hao Zhang, Kaichen Zhang, Yanwei Li, Ziwei Liu, and Chunyuan Li. Llava-onevision: Easy visual task transfer. arXiv preprint arXiv:2408.03326, 2024. 2

    2024 arXiv

  29. [37]

    Videochat: Chat-centric video understanding

    KunChang Li, Yinan He, Yi Wang, Yizhuo Li, Wenhai Wang, Ping Luo, Yali Wang, Limin Wang, and Yu Qiao. Videochat: Chat-centric video understanding. arXiv preprint arXiv:2305.06355, 2023. 3

    2023 arXiv

  30. [38]

    Segearth-ov: Towards traning-free open-vocabulary segmentation for remote sensing images

    Kaiyu Li, Ruixun Liu, Xiangyong Cao, Deyu Meng, and Zhi Wang. Segearth-ov: Towards traning-free open-vocabulary segmentation for remote sensing images. arXiv preprint arXiv:2410.01768, 2024. 6, 8

    2024 arXiv

  31. [39]

    Language-guided progressive attention for visual grounding in remote sensing images

    Ke Li, Di Wang, Haojie Xu, Haodi Zhong, and Cong Wang. Language-guided progressive attention for visual grounding in remote sensing images. IEEE Transactions on Geoscience and Remote Sensing, 2024. 6, 1, 5

    2024

  32. [40]

    Vrsbench: A versatile vision-language benchmark dataset for remote sens- ing image understanding

    Xiang Li, Jian Ding, and Mohamed Elhoseiny. Vrsbench: A versatile vision-language benchmark dataset for remote sens- ing image understanding. arXiv preprint arXiv:2406.12384,

    arXiv

  33. [41]

    Llama-vid: An image is worth 2 tokens in large language models

    Yanwei Li, Chengyao Wang, and Jiaya Jia. Llama-vid: An image is worth 2 tokens in large language models. In European Conference on Computer Vision, pages 323–340. Springer, 2025. 3

    2025

  34. [42]

    Unifiedmllm: Enabling unified representation for multi-modal multi-tasks with large language model

    Zhaowei Li, Wei Wang, Yiqing Cai, Xu Qi, Pengyu Wang, Dong Zhang, Hang Song, Botian Jiang, Zhida Huang, and Tao Wang. Unifiedmllm: Enabling unified representation for multi-modal multi-tasks with large language model. arXiv preprint arXiv:2408.02503, 2024. 2, 3

    2024 arXiv

  35. [43]

    Moe-llava: Mixture of experts for large vision- language models

    Bin Lin, Zhenyu Tang, Yang Ye, Jiaxi Cui, Bin Zhu, Peng Jin, Jinfa Huang, Junwu Zhang, Yatian Pang, Munan Ning, et al. Moe-llava: Mixture of experts for large vision- language models. arXiv preprint arXiv:2401.15947 , 2024. 2, 3

    2024 arXiv

  36. [44]

    Sphinx: The joint mixing of weights, tasks, and visual embeddings for multi-modal large language models

    Ziyi Lin, Chris Liu, Renrui Zhang, Peng Gao, Longtian Qiu, Han Xiao, Han Qiu, Chen Lin, Wenqi Shao, Keqin Chen, et al. Sphinx: The joint mixing of weights, tasks, and visual embeddings for multi-modal large language models. arXiv preprint arXiv:2311.07575, 2023. 2

    2023 arXiv

  37. [45]

    Deepseek-v2: A strong, economical, and efficient mixture-of-experts language model

    Aixin Liu, Bei Feng, Bin Wang, Bingxuan Wang, Bo Liu, Chenggang Zhao, Chengqi Dengr, Chong Ruan, Damai Dai, Daya Guo, et al. Deepseek-v2: A strong, economical, and efficient mixture-of-experts language model. arXiv preprint arXiv:2405.04434, 2024. 3

    2024 arXiv

  38. [46]

    Remote sensing image change captioning with dual-branch transformers: A new method and a large scale dataset

    Chenyang Liu, Rui Zhao, Hao Chen, Zhengxia Zou, and Zhenwei Shi. Remote sensing image change captioning with dual-branch transformers: A new method and a large scale dataset. IEEE Transactions on Geoscience and Remote Sens- ing, 60:1–20, 2022. 6

    2022

  39. [47]

    A decoupling paradigm with prompt learning for remote sensing image change captioning

    Chenyang Liu, Rui Zhao, Jianqi Chen, Zipeng Qi, Zhengxia Zou, and Zhenwei Shi. A decoupling paradigm with prompt learning for remote sensing image change captioning. IEEE Transactions on Geoscience and Remote Sensing, 2023. 6

    2023

  40. [48]

    Rscama: Remote sensing image change captioning with state space model

    Chenyang Liu, Keyan Chen, Bowen Chen, Haotian Zhang, Zhengxia Zou, and Zhenwei Shi. Rscama: Remote sensing image change captioning with state space model. IEEE Geo- science and Remote Sensing Letters, 2024. 6

    2024

  41. [49]

    Change-agent: Towards interactive comprehensive remote sensing change interpre- tation and analysis

    Chenyang Liu, Keyan Chen, Haotian Zhang, Zipeng Qi, Zhengxia Zou, and Zhenwei Shi. Change-agent: Towards interactive comprehensive remote sensing change interpre- tation and analysis. IEEE Transactions on Geoscience and Remote Sensing, 2024. 2, 3, 6, 8, 1, 5

    2024

  42. [50]

    Improved baselines with visual instruction tuning

    Haotian Liu, Chunyuan Li, Yuheng Li, and Yong Jae Lee. Improved baselines with visual instruction tuning. In Pro- ceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pages 26296–26306, 2024. 5

    2024

  43. [51]

    Visual instruction tuning

    Haotian Liu, Chunyuan Li, Qingyang Wu, and Yong Jae Lee. Visual instruction tuning. Advances in neural information processing systems, 36, 2024. 2, 3

    2024

  44. [52]

    Rsvqa: Visual question answering for remote sensing data

    Sylvain Lobry, Diego Marcos, Jesse Murray, and Devis Tuia. Rsvqa: Visual question answering for remote sensing data. IEEE Transactions on Geoscience and Remote Sensing , 58 (12):8555–8566, 2020. 5, 6, 8, 1

    2020

  45. [53]

    Exploring models and data for remote sensing im- age caption generation

    Xiaoqiang Lu, Binqiang Wang, Xiangtao Zheng, and Xue- long Li. Exploring models and data for remote sensing im- age caption generation. IEEE Transactions on Geoscience and Remote Sensing, 56(4):2183–2195, 2017. 1

    2017

  46. [54]

    Skysensegpt: A fine-grained in- struction tuning dataset and model for remote sensing vision- language understanding

    Junwei Luo, Zhen Pang, Yongjun Zhang, Tingzhu Wang, Linlin Wang, Bo Dang, Jiangwei Lao, Jian Wang, Jingdong Chen, Yihua Tan, et al. Skysensegpt: A fine-grained in- struction tuning dataset and model for remote sensing vision- language understanding. arXiv preprint arXiv:2406.10100,

    arXiv

  47. [55]

    Uavid: A semantic segmentation dataset for uav imagery

    Ye Lyu, George V osselman, Gui-Song Xia, Alper Yilmaz, and Michael Ying Yang. Uavid: A semantic segmentation dataset for uav imagery. ISPRS journal of photogrammetry and remote sensing, 165:108–119, 2020. 1, 5

    2020

  48. [56]

    Groma: Localized visual tokenization for grounding multimodal large language models

    Chuofan Ma, Yi Jiang, Jiannan Wu, Zehuan Yuan, and Xiao- juan Qi. Groma: Localized visual tokenization for grounding multimodal large language models. In European Conference on Computer Vision, pages 417–435. Springer, 2025. 2

    2025

  49. [57]

    Chartqa: A benchmark for question an- swering about charts with visual and logical reasoning.arXiv preprint arXiv:2203.10244, 2022

    Ahmed Masry, Do Xuan Long, Jia Qing Tan, Shafiq Joty, and Enamul Hoque. Chartqa: A benchmark for question an- swering about charts with visual and logical reasoning.arXiv preprint arXiv:2203.10244, 2022. 5

    2022 arXiv

  50. [58]

    Docvqa: A dataset for vqa on document images

    Minesh Mathew, Dimosthenis Karatzas, and CV Jawahar. Docvqa: A dataset for vqa on document images. In Proceed- ings of the IEEE/CVF winter conference on applications of computer vision, pages 2200–2209, 2021. 5

    2021

  51. [59]

    Semi-supervised semantic segmentation with high-and low- level consistency

    Sudhanshu Mittal, Maxim Tatarchenko, and Thomas Brox. Semi-supervised semantic segmentation with high-and low- level consistency. IEEE transactions on pattern analysis and machine intelligence, 43(4):1369–1379, 2019. 6

    2019

  52. [60]

    Lhrs-bot: Empowering remote sensing with vgi-enhanced large multimodal language model

    Dilxat Muhtar, Zhenshi Li, Feng Gu, Xueliang Zhang, and Pengfeng Xiao. Lhrs-bot: Empowering remote sensing with vgi-enhanced large multimodal language model. arXiv preprint arXiv:2402.02544, 2024. 2, 3, 5, 6, 1

    2024 arXiv

  53. [61]

    Cdchat: A large multimodal model for remote sensing change description

    Mubashir Noman, Noor Ahsan, Muzammal Naseer, Hisham Cholakkal, Rao Muhammad Anwer, Salman Khan, and Fa- had Shahbaz Khan. Cdchat: A large multimodal model for remote sensing change description. arXiv preprint arXiv:2409.16261, 2024. 2, 3

    2024 arXiv

  54. [62]

    H2rsvlm: Towards helpful and honest re- mote sensing large vision language model

    Chao Pang, Jiang Wu, Jiayu Li, Yi Liu, Jiaxing Sun, Wei- jia Li, Xingxing Weng, Shuai Wang, Litong Feng, Gui- Song Xia, et al. H2rsvlm: Towards helpful and honest re- mote sensing large vision language model. arXiv preprint arXiv:2403.20213, 2024. 2, 3, 5

    2024 arXiv

  55. [63]

    Semicdnet: A semisu- pervised convolutional neural network for change detection in high resolution remote-sensing images

    Daifeng Peng, Lorenzo Bruzzone, Yongjun Zhang, Haiyan Guan, Haiyong Ding, and Xu Huang. Semicdnet: A semisu- pervised convolutional neural network for change detection in high resolution remote-sensing images. IEEE Transac- tions on Geoscience and Remote Sensing, 59(7):5891–5906,

  56. [64]

    Improving language understanding by gener- ative pre-training

    Alec Radford. Improving language understanding by gener- ative pre-training. 2018. 2, 3

    2018

  57. [65]

    Language models are unsu- pervised multitask learners

    Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, Ilya Sutskever, et al. Language models are unsu- pervised multitask learners. OpenAI blog, 1(8):9, 2019. 2, 3

    2019

  58. [66]

    Learning transferable visual models from natural language supervi- sion

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, et al. Learning transferable visual models from natural language supervi- sion. In International conference on machine learning, ...

    2021

  59. [67]

    Exploring the limits of transfer learning with a unified text-to-text transformer

    Colin Raffel, Noam Shazeer, Adam Roberts, Katherine Lee, Sharan Narang, Michael Matena, Yanqi Zhou, Wei Li, and Peter J Liu. Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of machine learning research, 21(140):1–67, 2020. 2, 3

    2020

  60. [68]

    A lightweight transformer for remote sensing image change captioning

    Dongwei Sun, Yajie Bao, and Xiangyong Cao. A lightweight transformer for remote sensing image change captioning. arXiv preprint arXiv:2405.06598, 2024. 6

    2024 arXiv

  61. [69]

    Advent: Adversarial entropy min- imization for domain adaptation in semantic segmentation

    Tuan-Hung Vu, Himalaya Jain, Maxime Bucher, Matthieu Cord, and Patrick P ´erez. Advent: Adversarial entropy min- imization for domain adaptation in semantic segmentation. In Proceedings of the IEEE/CVF conference on computer vi- sion and pattern recognition, pages 2517–2526, 2019. 6

    2019

  62. [70]

    Sclip: Rethink- ing self-attention for dense vision-language inference

    Feng Wang, Jieru Mei, and Alan Yuille. Sclip: Rethink- ing self-attention for dense vision-language inference. In European Conference on Computer Vision, pages 315–332. Springer, 2025. 6

    2025

  63. [71]

    Loveda: A remote sensing land-cover dataset for domain adaptive semantic segmentation

    Junjue Wang, Zhuo Zheng, Ailong Ma, Xiaoyan Lu, and Yanfei Zhong. Loveda: A remote sensing land-cover dataset for domain adaptive semantic segmentation. arXiv preprint arXiv:2110.08733, 2021. 1, 5

    2021 arXiv

  64. [72]

    Reliable contrastive learning for semi- supervised change detection in remote sensing images.IEEE Transactions on Geoscience and Remote Sensing , 60:1–13,

    Jia-Xin Wang, Teng Li, Si-Bao Chen, Jin Tang, Bin Luo, and Richard C Wilson. Reliable contrastive learning for semi- supervised change detection in remote sensing images.IEEE Transactions on Geoscience and Remote Sensing , 60:1–13,

  65. [73]

    Cogvlm: Visual expert for pretrained language models

    Weihan Wang, Qingsong Lv, Wenmeng Yu, Wenyi Hong, Ji Qi, Yan Wang, Junhui Ji, Zhuoyi Yang, Lei Zhao, Xixuan Song, et al. Cogvlm: Visual expert for pretrained language models. arXiv preprint arXiv:2311.03079, 2023. 5

    2023 arXiv

  66. [74]

    The all-seeing project v2: To- wards general relation comprehension of the open world

    Weiyun Wang, Yiming Ren, Haowen Luo, Tiantong Li, Chenxiang Yan, Zhe Chen, Wenhai Wang, Qingyun Li, Lewei Lu, Xizhou Zhu, et al. The all-seeing project v2: To- wards general relation comprehension of the open world. In European Conference on Computer Vision, pages 471–490. Spr...

    2025

  67. [75]

    Chain-of-thought prompting elicits reasoning in large lan- guage models

    Jason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma, Fei Xia, Ed Chi, Quoc V Le, Denny Zhou, et al. Chain-of-thought prompting elicits reasoning in large lan- guage models. Advances in neural information processing systems, 35:24824–24837, 2022. 8

    2022

  68. [76]

    Visionllm v2: An end-to-end general- ist multimodal large language model for hundreds of vision- language tasks

    Jiannan Wu, Muyan Zhong, Sen Xing, Zeqiang Lai, Zhaoyang Liu, Wenhai Wang, Zhe Chen, Xizhou Zhu, Lewei Lu, Tong Lu, et al. Visionllm v2: An end-to-end general- ist multimodal large language model for hundreds of vision- language tasks. arXiv preprint arXiv:2406.08394, 2024. 2, 3

    2024 arXiv

  69. [77]

    Aid: A benchmark data set for performance evaluation of aerial scene classification

    Gui-Song Xia, Jingwen Hu, Fan Hu, Baoguang Shi, Xiang Bai, Yanfei Zhong, Liangpei Zhang, and Xiaoqiang Lu. Aid: A benchmark data set for performance evaluation of aerial scene classification. IEEE Transactions on Geoscience and Remote Sensing, 55(7):3965–3981, 2017. 7

    2017

  70. [78]

    Openearthmap: A benchmark dataset for global high-resolution land cover mapping

    Junshi Xia, Naoto Yokoya, Bruno Adriano, and Clifford Broni-Bediako. Openearthmap: A benchmark dataset for global high-resolution land cover mapping. In Proceed- ings of the IEEE/CVF Winter Conference on Applications of Computer Vision, pages 6254–6264, 2023. 6, 1, 5

    2023

  71. [79]

    Florence-2: Advancing a unified representation for a variety of vision tasks

    Bin Xiao, Haiping Wu, Weijian Xu, Xiyang Dai, Houdong Hu, Yumao Lu, Michael Zeng, Ce Liu, and Lu Yuan. Florence-2: Advancing a unified representation for a variety of vision tasks. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition , pages 481...

    2024

  72. [80]

    Qa-lora: Quantization-aware low-rank adaptation of large language models

    Yuhui Xu, Lingxi Xie, Xiaotao Gu, Xin Chen, Heng Chang, Hengheng Zhang, Zhengsu Chen, Xiaopeng Zhang, and Qi Tian. Qa-lora: Quantization-aware low-rank adaptation of large language models. arXiv preprint arXiv:2309.14717 ,

    arXiv

  73. [81]

    Multiinstruct: Im- proving multi-modal zero-shot learning via instruction tun- ing

    Zhiyang Xu, Ying Shen, and Lifu Huang. Multiinstruct: Im- proving multi-modal zero-shot learning via instruction tun- ing. arXiv preprint arXiv:2212.10773, 2022. 2

    2022 arXiv

  74. [82]

    Bag-of-visual-words and spa- tial extensions for land-use classification

    Yi Yang and Shawn Newsam. Bag-of-visual-words and spa- tial extensions for land-use classification. In Proceedings of the 18th SIGSPATIAL international conference on advances in geographic information systems, pages 270–279, 2010. 1, 5

    2010

  75. [83]

    Minicpm-v: A gpt-4v level mllm on your phone

    Yuan Yao, Tianyu Yu, Ao Zhang, Chongyi Wang, Junbo Cui, Hongji Zhu, Tianchi Cai, Haoyu Li, Weilin Zhao, Zhihui He, et al. Minicpm-v: A gpt-4v level mllm on your phone. arXiv preprint arXiv:2408.01800, 2024. 2, 6

    2024 arXiv

  76. [84]

    Exploring a fine-grained multiscale method for cross-modal remote sensing image re- trieval

    Zhiqiang Yuan, Wenkai Zhang, Kun Fu, Xuan Li, Chubo Deng, Hongqi Wang, and Xian Sun. Exploring a fine-grained multiscale method for cross-modal remote sensing image re- trieval. arXiv preprint arXiv:2204.09868, 2022. 1, 5

    2022 arXiv

  77. [85]

    Sigmoid loss for language image pre-training

    Xiaohua Zhai, Basil Mustafa, Alexander Kolesnikov, and Lucas Beyer. Sigmoid loss for language image pre-training. In Proceedings of the IEEE/CVF International Conference on Computer Vision, pages 11975–11986, 2023. 2

    2023

  78. [86]

    Rsvg: Exploring data and models for visual grounding on remote sensing data

    Yang Zhan, Zhitong Xiong, and Yuan Yuan. Rsvg: Exploring data and models for visual grounding on remote sensing data. IEEE Transactions on Geoscience and Remote Sensing , 61: 1–13, 2023. 5, 6, 8, 1

    2023

  79. [87]

    Skyeyegpt: Unifying remote sensing vision-language tasks via instruc- tion tuning with large language model

    Yang Zhan, Zhitong Xiong, and Yuan Yuan. Skyeyegpt: Unifying remote sensing vision-language tasks via instruc- tion tuning with large language model. arXiv preprint arXiv:2401.09712, 2024. 3, 5

    2024 arXiv

  80. [88]

    Next-chat: An lmm for chat, detection and segmentation

    Ao Zhang, Liming Zhao, Chen-Wei Xie, Yun Zheng, Wei Ji, and Tat-Seng Chua. Next-chat: An lmm for chat, detection and segmentation. arXiv preprint arXiv:2311.04498, 2023. 2

    2023 arXiv

  81. [89]

    Llava-grounding: Grounded visual chat with large multimodal models

    Hao Zhang, Hongyang Li, Feng Li, Tianhe Ren, Xueyan Zou, Shilong Liu, Shijia Huang, Jianfeng Gao, Chunyuan Li, Jainwei Yang, et al. Llava-grounding: Grounded visual chat with large multimodal models. In European Conference on Computer Vision, pages 19–35. Springer, 2025. 2

    2025

  82. [90]

    Internlm-xcomposer-2.5: A versatile large vision language model supporting long-contextual in- put and output

    Pan Zhang, Xiaoyi Dong, Yuhang Zang, Yuhang Cao, Rui Qian, Lin Chen, Qipeng Guo, Haodong Duan, Bin Wang, Linke Ouyang, et al. Internlm-xcomposer-2.5: A versatile large vision language model supporting long-contextual in- put and output. arXiv preprint arXiv:2407.03320, 2024. 5

    2024 arXiv

  83. [91]

    Omg-llava: Bridging image-level, object-level, pixel-level reasoning and understanding

    Tao Zhang, Xiangtai Li, Hao Fei, Haobo Yuan, Shengqiong Wu, Shunping Ji, Chen Change Loy, and Shuicheng Yan. Omg-llava: Bridging image-level, object-level, pixel-level reasoning and understanding. arXiv preprint arXiv:2406.19389, 2024. 2

    2024 arXiv

  84. [92]

    Robust mixture-of-expert training for convolu- tional neural networks

    Yihua Zhang, Ruisi Cai, Tianlong Chen, Guanhua Zhang, Huan Zhang, Pin-Yu Chen, Shiyu Chang, Zhangyang Wang, and Sijia Liu. Robust mixture-of-expert training for convolu- tional neural networks. In Proceedings of the IEEE/CVF In- ternational Conference on Computer Vision, pages 90–101,

  85. [93]

    Rs5m and georsclip: A large scale vision-language dataset and a large vision-language model for remote sensing

    Zilun Zhang, Tiancheng Zhao, Yulong Guo, and Jianwei Yin. Rs5m and georsclip: A large scale vision-language dataset and a large vision-language model for remote sensing. IEEE Transactions on Geoscience and Remote Sensing, 2024. 2

    2024

  86. [94]

    Extract free dense labels from clip

    Chong Zhou, Chen Change Loy, and Bo Dai. Extract free dense labels from clip. In European Conference on Com- puter Vision, pages 696–712. Springer, 2022. 6, 8

    2022

  87. [95]

    Single- stream extractor network with contrastive pre-training for re- mote sensing change captioning

    Qing Zhou, Junyu Gao, Yuan Yuan, and Qi Wang. Single- stream extractor network with contrastive pre-training for re- mote sensing change captioning. IEEE Transactions on Geo- science and Remote Sensing, 2024. 6

    2024

  88. [96]

    Minigpt-4: Enhancing vision-language understanding with advanced large language models

    Deyao Zhu, Jun Chen, Xiaoqian Shen, Xiang Li, and Mo- hamed Elhoseiny. Minigpt-4: Enhancing vision-language understanding with advanced large language models. arXiv preprint arXiv:2304.10592, 2023. 2, 3, 5

    2023 arXiv

  89. [97]

    Bag-of-visual-words scene classifier with local and global features for high spatial resolution remote sensing imagery

    Qiqi Zhu, Yanfei Zhong, Bei Zhao, Gui-Song Xia, and Liangpei Zhang. Bag-of-visual-words scene classifier with local and global features for high spatial resolution remote sensing imagery. IEEE Geoscience and Remote Sensing Let- ters, 13(6):747–751, 2016. 7 RSUniVLM: A Unified ...

    2016

  90. [98]

    As shown in Tab

    Details of Dataset We construct a universal multi-task instruction-following dataset by leveraging a variety of public RS datasets, to- gether with several existing RS and general instruction datasets. As shown in Tab. 10, we collect and reorganize 15 public datasets across 6 ...

  91. [99]

    Task Tokens To enable RSUniVLM to distinguish different tasks more flexibly, we insert a task-specific token at the beginning of the text prompt, following the previous works [32, 60]. Spe- cially, we add token [CAP], [CLS], [VQA], [VG], [REF], [SEG] and [CCD] for image captio...

  92. [100]

    we employ the AdamW optimizer along with a learning rate of 0.00005, warmup ratio of 0.03 and weight decay of 0

    Hyperparameter We trained all models on 4 Nvidia A40 GPUs (40GB). we employ the AdamW optimizer along with a learning rate of 0.00005, warmup ratio of 0.03 and weight decay of 0. Both two training stages last for 1 epoch, with a batch size of 4. See Tab. 9 for detailed trainin...

  93. [101]

    Stage1 Stage2 Trainable Parts Image Encoder, Connector, LLM G-MoE Optimizer AdamW AdamW Learning Rate 1e-5 1e-5 Weight Decay 0

    Prompt for Evaluation In this section, we describe the prompts we use when eval- uating each task in detail. Stage1 Stage2 Trainable Parts Image Encoder, Connector, LLM G-MoE Optimizer AdamW AdamW Learning Rate 1e-5 1e-5 Weight Decay 0. 0. Warmup Ratio 0.03 0.03 Scheduler Cosi...

  94. [102]

    More Qualitative Results We present more qualitative results in Fig. 4, Fig. 5, Fig. 6, which presents the visual grounding, semantic segmenta- Change Captioning & Change Detection [SEG] Please segment the building area in the images that have changed in the second image. [SEG...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.