Pith. sign in

REVIEW 3 major objections 6 minor 4 cited by

CCExpert: Advancing MLLM Capability in Remote Sensing Change Captioning with Difference-Aware Integration and a Foundational Dataset

T0 review · 3 major / 6 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read The paper claims CCExpert reaches 81.80 on the LEVIR-CC change-captioning benchmark by injecting multi-scale difference features into a multimodal language model and continuing pretraining on a 200,000-pair dataset.

desk verdict The dataset and module are real, but the SOTA claim is compromised by a likely train/test leak and test-set hyperparameter selection. read the letter →

arxiv 2411.11360 v1 pith:A5PSI4BX submitted 2024-11-18 cs.CV

Zhiming Wang , Mingze Wang , Sheng Xu , Yanjing Li , Baochang Zhang This is my paper · ORCID
classification cs.CV
keywords remotesensingimagechangecaptioningmultimodallargelanguagemodeldifference-awareintegrationmoduleCC-FoundationdatasetLEVIR-CCbenchmarkcontinuedpretrainingmulti-scalefeaturefusion
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 tries to establish that a multimodal large language model can become the best remote sensing change captioner without overhauling its core architecture, by explicitly feeding it multi-scale differences between the two images and by continuing pretraining on a large domain dataset. The authors build CCExpert on the LLaVA-OneVision base, add a Difference-aware Integration Module that extracts change features from several vision-encoder layers and weights them into the original image features, and assemble CC-Foundation: 200,000 image pairs with 1.2 million captions. They report a composite score of $S^*_m = 81.80$ on LEVIR-CC, roughly two points above the previous best listed method, and attribute the gain to the dataset and the module. If true, this makes data-driven continued pretraining plus targeted difference injection a practical recipe for adapting general multimodal models to change understanding.

What carries the argument

The carrying mechanism is the Difference-aware Integration Module, which has two parts. A Diff Expert takes features from vision-transformer layers at positions $-11$, $-8$, $-5$, and $-2$, builds initial difference maps by gating each temporal feature with the other through concatenation and a sigmoid, refines those maps through two Change Aware Transformer Layers using self-attention and cross-attention, and injects the refined difference back into both image feature streams. An Adaptive Adjustment submodule computes per-scale weights, sums the weighted multi-scale features, and adds them residually to the original features. This module is what carries the argument that explicit multi-scale difference information, rather than a redesigned vision encoder or task head, lets the pretrained language model reason about changes. The other load-bearing pieces are the CC-Foundation dataset, built from existing change captioning and change detection datasets with LLM refinement, GPT-4o-generated captions from change masks, and expert annotations; and the three-stage training schedule that pretrains the new module, then unfreezes the whole model, then fine-tunes on domain data.

What would settle it

A direct check is to inspect the released training lists for LEVIR-CC test filenames, then rerun the CCExpert-7B recipe from scratch with the test split excluded and with the multi-scale layers and transformer count chosen on a held-out validation split instead of the test table; if the composite score falls to or below the 80.19 of the unmodified base model, the claimed gain is an artifact of test-set exposure.

Watch

Extended reading notes

Core claim

The central claim is that CCExpert sets a new state of the art on the LEVIR-CC remote sensing image change captioning benchmark, reaching $S^*_m = 81.80$, where $S^*_m$ is the average of BLEU-4, ROUGE-L, METEOR, and CIDEr-D. The unmodified base model reaches 80.19 with supervised fine-tuning alone, and the previous best listed method reaches about 79.6. The authors attribute the gain to three coordinated choices: keep the standard multimodal model pathway intact, inject multi-scale bi-temporal difference features into the visual tokens before the projector, and continue pretraining on CC-Foundation before domain fine-tuning. Ablations show that the continued pretraining adds about 1.1 points and the Difference-aware Integration module adds about 0.5 points in the 7B setting, so the two together move the model from 80.19 to 81.80.

Load-bearing premise

The load-bearing premise is that the LEVIR-CC test partition is absent from the CC-Foundation pretraining data and from the ablation hyperparameter choices, so the reported 81.80 measures genuine generalization rather than memorization.

Editorial extensions

If this is right

  • If the reported score is robust, a standard multimodal language model can be converted into a top change captioner by feature-level difference injection and continued pretraining, without replacing the vision encoder or adding task heads.
  • The CC-Foundation dataset, if released, becomes a reusable resource for pretraining and evaluating remote sensing change captioning models, including the expert-annotated SECOND-derived split the authors propose as a future benchmark.
  • The three-stage schedule offers a template for adding lightweight modules to multimodal language models without destructively fine-tuning the base weights.
  • The 0.5B variant reaching 80.99 suggests most of the gain does not require a 7B decoder, so the recipe may transfer to smaller deployable models.

Reading between the lines

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

  • The paper's own tables list all 10,000 LEVIR-CC image pairs as pretraining data and select the multi-scale layer set and transformer count from LEVIR-CC test metrics; unless the test split was withheld, part of the reported gain could be memorization rather than understanding, and the paper does not state that it was withheld.
  • Several ablated choices move different metrics in different directions, so the reported ranking is sensitive to the choice of the composite $S^*_m$ as the selection criterion rather than any single captioning metric.
  • Because CC-Foundation captions are partly generated by GPT-4o from change-mask prompts, models trained on it may inherit the annotation style and biases of that generator; the paper's own limitation section acknowledges possible errors in these annotations.
  • The manuscript labels itself as work in progress and evaluates on a single benchmark, so the stated lead should be read as a status report pending independent reproduction on additional datasets.
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 / 6 minor

Summary. The paper proposes CCExpert, a remote sensing image change captioning (RSICC) model built on LLaVA-OneVision, with three components: a Difference-aware Integration Module that injects multi-scale change features into the original image representation, a large continued-pretraining corpus called CC-Foundation (200k image pairs, 1.2M captions), and a three-stage training schedule. The authors report S*_m = 81.80 on the LEVIR-CC benchmark, claiming a large improvement over previous state-of-the-art methods, and ablate the base model, the continued-pretraining data, the difference module, the multi-scale layer set, and the number of Change Aware Transformer layers.

Significance. If the reported results were obtained under a clean evaluation protocol, CCExpert would be a meaningful advance: it demonstrates a modular way to inject change information into an MLLM without disrupting the pretrained architecture, and the CC-Foundation dataset (even if only partially released) would be a useful resource for RSICC. The consistent gains across the 0.5B and 7B variants are suggestive. However, the current evaluation protocol does not support the headline claim: the manuscript does not rule out that the LEVIR-CC test split was used in continued pretraining, and hyperparameters are explicitly selected on the LEVIR-CC test set. The reported 81.80 therefore cannot be interpreted as an unbiased estimate of generalization. The significance of the contribution is currently not established.

major comments (3)
  1. [Section IV-A1 and IV-A4, Table I] The manuscript never states whether the LEVIR-CC test split is excluded from the CC-Foundation training corpus. Table I lists LEVIR-CC (10k image pairs) as a component of CC-Foundation, and Section IV-A4 says 'LEVIR-CC, SYSU-CD, and SECOND datasets are upsampled threefold' for training, with no statement that the standard train/test split was respected. If any LEVIR-CC test pair was in the training data, the reported S*_m = 81.80 on LEVIR-CC is a leakage artifact and the state-of-the-art claim is invalid. The authors must state explicitly which LEVIR-CC split is used for training; if the full dataset was used, the experiments must be redone with the test split excluded.
  2. [Section IV-C4 and IV-C5, Tables VII and VIII] The captions of Tables VII and VIII say 'Performances are validated on LEVIR-CC test set,' and the multi-scale layer set {-2,-5,-8,-11} and the number of Change Aware Transformer Layers (2) are selected by comparing test-set scores. This is test-set model selection, which introduces an optimistic bias into the reported S*_m = 81.80. The authors should split LEVIR-CC into train/validation/test (or use cross-validation) for hyperparameter selection, and report test-set performance only for the final chosen configuration. Without this, the reported improvements cannot be taken as evidence of generalization.
  3. [Section IV-C, Tables IV-VI] All experiments appear to be single runs without error bars or significance tests. The main reported gains are small (0.49 to 1.12 in S*_m), and in MLLM fine-tuning such differences can easily fall within run-to-run variance. The authors should report mean and standard deviation over at least three random seeds for the main baseline, the continued-pretraining ablation, and the final CCExpert model, so the reader can assess whether the differences are meaningful.
minor comments (6)
  1. [Section III-B1] The citation to Semantic-CC appears as '[ ? ]'; please fill in the missing reference.
  2. [Section IV-C2, Table V] The text states 'All evaluation metrics improved' after continued pretraining, but for CCExpert-7B, BLEU-1 decreases from 86.37 to 86.35. Please correct either the sentence or the table.
  3. [Section IV-A4] The vision encoder is referred to as both 'siglip-so400m' and 'Siglip-400m' in different places; please standardize the naming.
  4. [Table VI caption] There is a typo in the caption: 'dfifference' should be 'difference'.
  5. [Section IV-C4] The sentence 'Based on these results, we selected {-2,-5,-8,-11} as the optimal layer combination' would be clearer if it specified that the selection was made on a validation set, once the evaluation protocol is corrected.
  6. [Title page] The header contains the note 'This work is currently in progress (WIP), with ongoing development and refinement.' For a journal submission, this note should be removed.

Circularity Check

2 steps flagged · score 8.0 of 10

The LEVIR-CC evaluation set is included in CC-Foundation training and used for hyperparameter selection, so the reported S*_m=81.80 is a fitted value rather than a prediction.

  1. fitted input called prediction [Section IV-A1 (CC-Foundation Dataset, Table I) and Section IV-A4 (Implementation Details)]
    "First, we included as many open-source datasets in this field as possible, such as CLVER-Change [55], ImageEdit [56], Spot-the-diff [57], stvchrono [58], Vismin [59], and LEVIR-CC [4]. ... LEVIR-CC, SYSU-CD, and SECOND datasets are upsampled threefold to increase the proportion of remote sensing images in training."

    Table I lists all 10k LEVIR-CC image pairs as part of CC-Foundation pretraining, and Section IV-A4 states that LEVIR-CC is upsampled threefold in training. The evaluation section then states 'we selected the LEVIR-CC dataset as the test set' and reports S*_m=81.80. The paper never states that the standard train/test split is respected or that LEVIR-CC test pairs are withheld from CC-Foundation. If any test pair appears in training, the reported improvement over prior methods is a leakage artifact; the claim that the method 'surpasses' baselines is then an artifact of training on the benchmark rather than a prediction.

  2. fitted input called prediction [Section IV-C4 and IV-C5, Table VII and Table VIII captions]
    "Performances are validated on LEVIR-CC test set. ... Based on these results, we selected {−2, −5, −8, −11} as the optimal layer combination for multi-scale feature input. ... When using two Change Aware Transformer Layers, the model achieved optimal performance across all metrics, with S∗m reaching 81.80."

    The number of Change Aware Transformer layers and the multi-scale feature layer set are chosen by comparing S*_m on the LEVIR-CC test set itself. The final reported 81.80 is the result of the configuration that maximized the test metric. Selecting hyperparameters directly on the test set makes the reported performance an optimistically biased fit to that benchmark; the 'optimal' settings and the resulting SOTA score are therefore not independent evidence for the method's effectiveness.

full rationale

The paper's central claim is that CCExpert achieves S*_m=81.80 on LEVIR-CC and 'significantly surpasses' prior methods. This claim depends on two conditions that the manuscript fails to establish and, on its face, contradicts: (1) that the LEVIR-CC test split is excluded from CC-Foundation pretraining and from Stage-III fine-tuning, and (2) that hyperparameters are not selected using test labels. Instead, Table I explicitly lists LEVIR-CC (10k image pairs) as part of CC-Foundation, and Section IV-A4 says LEVIR-CC is upsampled threefold in training; no holdout split is mentioned anywhere. Tables VII and VIII then select the multi-scale layer set and the number of transformer layers by their scores on the LEVIR-CC test set. The reported S*_m=81.80 is therefore best interpreted as a value fitted to the evaluation benchmark, not a clean prediction. This is not a case of self-citation or ansatz-smuggling; the circularity is a test-contamination / fitted-input problem. If the authors had explicitly documented an LEVIR-CC train/test split with test images excluded from all training and model selection, the result could have been a legitimate prediction; as written, the derivation chain is broken.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim is empirical and rests on dataset-split assumptions, test-based model selection, and synthetic annotation quality rather than on a formal derivation. The ledger lists the choices that the paper does not independently justify.

free parameters (5)
  • multi-scale layer set {-2,-5,-8,-11} = chosen from {-2}, {-2,-4,-6,-8}, {-2,-5,-8,-11}, {-2,-6,-10,-14}
    Selected as the best configuration on the LEVIR-CC test set in Table VII.
  • number of Change Aware Transformer layers = 2
    Selected by comparing 1, 2, and 3 layers using test-set S*_m in Table VIII.
  • base MLLM and model size = LLaVA-OneVision-7B
    Chosen because it scored highest on the LEVIR-CC test set in Table IV.
  • LEVIR-CC inclusion in CC-Foundation = 10k image pairs, the full dataset as listed
    Table I lists all LEVIR-CC pairs as pretraining data; if the test split is included, the headline score is fitted to the test set. No exclusion is stated.
  • remote sensing data upsampling factor = 3x
    Section IV-A4 upsamples LEVIR-CC, SYSU-CD, and SECOND threefold in training, with no ablation for this choice.
assumptions (4)
  • ad hoc to paper The LEVIR-CC test split is not included in the CC-Foundation pretraining corpus.
    Required for the reported test score to be meaningful. The paper never states this exclusion, and Table I plus Section IV-A4 imply all LEVIR-CC pairs are used in training.
  • domain assumption Hyperparameter choices can be validated on the test set without biasing the reported result.
    Tables VII and VIII select the layer set and layer count using test-set S*_m; this is only valid if test scores are not used for model selection, which is not the protocol described.
  • domain assumption GPT-4o-generated captions and expert-refined annotations are accurate enough to serve as a training signal.
    CC-Foundation includes 72k and 65k captions generated through GPT-4o with mask prompts, and the discussion in Section IV-D acknowledges the generated annotations may contain errors.
  • domain assumption Three-stage continued pretraining preserves the base MLLM's knowledge and does not corrupt the pretrained representations.
    Section III-E2 motivates this staged strategy, but the paper only reports final scores and does not analyze knowledge retention directly.

how reviews work

0 comments
Cite this review

Pith. "Pith review of CCExpert: Advancing MLLM Capability in Remote Sensing Change Captioning with Difference-Aware Integration and a Foundational Dataset." pith.science (2026). https://pith.science/paper/A5PSI4BX

@misc{pith2026241111360,
  author       = {Pith},
  title        = {Pith review of: CCExpert: Advancing MLLM Capability in Remote Sensing Change Captioning with Difference-Aware Integration and a Foundational Dataset},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/A5PSI4BX}},
  note         = {Machine review of arXiv:2411.11360}
}
abstract

Remote Sensing Image Change Captioning (RSICC) aims to generate natural language descriptions of surface changes between multi-temporal remote sensing images, detailing the categories, locations, and dynamics of changed objects (e.g., additions or disappearances). Many current methods attempt to leverage the long-sequence understanding and reasoning capabilities of multimodal large language models (MLLMs) for this task. However, without comprehensive data support, these approaches often alter the essential feature transmission pathways of MLLMs, disrupting the intrinsic knowledge within the models and limiting their potential in RSICC. In this paper, we propose a novel model, CCExpert, based on a new, advanced multimodal large model framework. Firstly, we design a difference-aware integration module to capture multi-scale differences between bi-temporal images and incorporate them into the original image context, thereby enhancing the signal-to-noise ratio of differential features. Secondly, we constructed a high-quality, diversified dataset called CC-Foundation, containing 200,000 image pairs and 1.2 million captions, to provide substantial data support for continue pretraining in this domain. Lastly, we employed a three-stage progressive training process to ensure the deep integration of the difference-aware integration module with the pretrained MLLM. CCExpert achieved a notable performance of $S^*_m=81.80$ on the LEVIR-CC benchmark, significantly surpassing previous state-of-the-art methods. The code and part of the dataset will soon be open-sourced at https://github.com/Meize0729/CCExpert.

Figures

Figures reproduced from arXiv: 2411.11360 by the authors.

Figure 1
Figure 1. Basic components of the proposed MLLM-based [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. The overall architecture, component details, and training strategy of CCExpert. Model Details(A): Adaptive Adjustment [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Visualization of the CC-Foundation Dataset. [PITH_FULL_IMAGE:figures/full_fig_p008_3.png] view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: Visual comparisons of the proposed method with other state-of-the-art methods. [PITH_FULL_IMAGE:figures/full_fig_p011_4.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. LongEarth-R1: Benchmarking and Aligning Vision-Language Models for Long-Horizon Earth Observation Reasoning

    cs.AI 2026-08 conditional novelty 6.0 of 10

    LongEarth-R1, a vision-language model trained with sequence-aware supervision and spatiotemporal GRPO, sets new state-of-the-art results on all 12 tasks of the new LongEarth-Bench for long-horizon Earth observation reasoning.

  2. EchoChange: A Diffusion Language Model with Dual Pass Remasking for Factual Remote Sensing Disaster Change Captioning

    cs.AI 2026-08 conditional novelty 6.0 of 10

    EchoChange generates remote sensing disaster captions by iterative masked-token denoising with dual-pass remasking, and reports large metric gains over autoregressive baselines on RSCC.

  3. Robust Change Captioning in Remote Sensing: SECOND-CC Dataset and MModalCC Framework

    cs.CV 2025-01 conditional novelty 6.0 of 10

    A new dataset and a multimodal attention model improve remote sensing change captioning, but only when ground-truth semantic maps are supplied as input.

  4. Change Captioning in Remote Sensing: Evolution to SAT-Cap -- A Single-Stage Transformer Approach

    cs.CV 2025-01 conditional novelty 5.0 of 10

    SAT-Cap, a single-stage transformer with spatial-channel attention and cosine-similarity fusion, achieves state-of-the-art CIDEr scores of 140.23% on LEVIR-CC and 97.74% on DUBAI-CCD for remote sensing change captioning.

Reference graph

Works this paper leans on

63 extracted references · 26 canonical work pages · cited by 4 Pith papers

  1. [1]

    Semantic- cc: Boosting remote sensing image change captioning via foundational knowledge and semantic guidance,

    Y . Zhu, L. Li, K. Chen, C. Liu, F. Zhou, and Z. Shi, “Semantic- cc: Boosting remote sensing image change captioning via foundational knowledge and semantic guidance,” arXiv preprint arXiv:2407.14032, 2024. JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 13

    arXiv 2024

  2. [2]

    Diffusion-rscc: Diffusion probabilistic model for change captioning in remote sensing images,

    X. Yu, Y . Li, and J. Ma, “Diffusion-rscc: Diffusion probabilistic model for change captioning in remote sensing images,” arXiv preprint arXiv:2405.12875, 2024

    arXiv 2024

  3. [3]

    Rscama: Remote sensing image change captioning with state space model,

    C. Liu, K. Chen, B. Chen, H. Zhang, Z. Zou, and Z. Shi, “Rscama: Remote sensing image change captioning with state space model,” IEEE Geoscience and Remote Sensing Letters , vol. 21, pp. 1–5, 2024

    2024

  4. [4]

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

    C. Liu, R. Zhao, H. Chen, Z. Zou, and Z. Shi, “Remote sensing image change captioning with dual-branch transformers: A new method and a large scale dataset,” IEEE Transactions on Geoscience and Remote Sensing , vol. 60, pp. 1–20, 2022

    2022

  5. [5]

    Progressive scale- aware network for remote sensing image change captioning,

    C. Liu, J. Yang, Z. Qi, Z. Zou, and Z. Shi, “Progressive scale- aware network for remote sensing image change captioning,” in IGARSS 2023-2023 IEEE International Geoscience and Remote Sensing Symposium . IEEE, 2023, pp. 6668–6671

    work page 2023

  6. [6]

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

    C. Liu, R. Zhao, J. Chen, Z. Qi, Z. Zou, and Z. Shi, “A decoupling paradigm with prompt learning for remote sensing image change captioning,” IEEE Transactions on Geoscience and Remote Sensing , vol. 61, pp. 1–18, 2023

    work page 2023

  7. [7]

    Single-stream ex- tractor network with contrastive pre-training for remote-sensing change captioning,

    Q. Zhou, J. Gao, Y . Yuan, and Q. Wang, “Single-stream ex- tractor network with contrastive pre-training for remote-sensing change captioning,” IEEE Transactions on Geoscience and Remote Sensing , vol. 62, pp. 1–14, 2024

    work page 2024

  8. [8]

    A lightweight transformer for remote sensing image change captioning,

    D. Sun, Y . Bao, and X. Cao, “A lightweight transformer for remote sensing image change captioning,” arXiv preprint arXiv:2405.06598, 2024

    arXiv 2024

Show all 63 references
  1. [9]

    Pixel-level change detection pseudo-label learning for remote sensing change captioning,

    C. Liu, K. Chen, Z. Qi, Z. Liu, H. Zhang, Z. Zou, and Z. Shi, “Pixel-level change detection pseudo-label learning for remote sensing change captioning,” in IGARSS 2024-2024 IEEE Inter- national Geoscience and Remote Sensing Symposium . IEEE, 2024, pp. 8405–8408

    2024

  2. [10]

    Segment anything,

    A. Kirillov, E. Mintun, N. Ravi, H. Mao, C. Rolland, L. Gustafson, T. Xiao, S. Whitehead, A. C. Berg, W.-Y . Lo et al. , “Segment anything,” arXiv preprint arXiv:2304.02643 , 2023

    2023 arXiv

  3. [11]

    Rsprompter: Learning to prompt for remote sensing instance segmentation based on visual foundation model,

    K. Chen, C. Liu, H. Chen, H. Zhang, W. Li, Z. Zou, and Z. Shi, “Rsprompter: Learning to prompt for remote sensing instance segmentation based on visual foundation model,” IEEE Transactions on Geoscience and Remote Sensing , 2024

    2024

  4. [12]

    Rsbuilding: Toward general remote sensing image building extraction and change detection with foundation model,

    M. Wang, L. Su, C. Yan, S. Xu, P. Yuan, X. Jiang, and B. Zhang, “Rsbuilding: Toward general remote sensing image building extraction and change detection with foundation model,” IEEE Transactions on Geoscience and Remote Sensing , vol. 62, pp. 1–17, 2024

    2024

  5. [13]

    Qwen2 technical report,

    A. Yang, B. Yang, B. Hui, B. Zheng, B. Yu, C. Zhou, C. Li, C. Li, D. Liu, F. Huang et al., “Qwen2 technical report,” arXiv preprint arXiv:2407.10671, 2024

    2024 arXiv

  6. [14]

    Gpt-4 technical report,

    J. Achiam, S. Adler, S. Agarwal, L. Ahmad, I. Akkaya, F. L. Aleman, D. Almeida, J. Altenschmidt, S. Altman, S. Anadkat et al. , “Gpt-4 technical report,” arXiv preprint arXiv:2303.08774, 2023

    2023 arXiv

  7. [15]

    Learning transferable visual models from natural language supervision,

    A. Radford, J. W. Kim, C. Hallacy, A. Ramesh, G. Goh, S. Agarwal, G. Sastry, A. Askell, P. Mishkin, J. Clark et al. , “Learning transferable visual models from natural language supervision,” in International conference on machine learning . PMLR, 2021, pp. 8748–8763

    2021

  8. [16]

    Qwen-vl: A versatile vision-language model for understanding, localization, text reading, and be- yond,

    J. Bai, S. Bai, S. Yang, S. Wang, S. Tan, P. Wang, J. Lin, C. Zhou, and J. Zhou, “Qwen-vl: A versatile vision-language model for understanding, localization, text reading, and be- yond,” arXiv preprint arXiv:2308.12966 , 2023

    2023 arXiv

  9. [17]

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

    D. Zhu, J. Chen, X. Shen, X. Li, and M. Elhoseiny, “Minigpt-4: Enhancing vision-language understanding with advanced large language models,” arXiv preprint arXiv:2304.10592 , 2023

    2023 arXiv

  10. [18]

    Visual instruction tuning,

    H. Liu, C. Li, Q. Wu, and Y . J. Lee, “Visual instruction tuning,” 2023

    2023

  11. [19]

    Llava-onevision: Easy visual task transfer,

    B. Li, Y . Zhang, D. Guo, R. Zhang, F. Li, H. Zhang, K. Zhang, Y . Li, Z. Liu, and C. Li, “Llava-onevision: Easy visual task transfer,” arXiv preprint arXiv:2408.03326 , 2024

    2024 arXiv

  12. [20]

    Llava-next: Improved reasoning, ocr, and world knowledge,

    H. Liu, C. Li, Y . Li, B. Li, Y . Zhang, S. Shen, and Y . J. Lee, “Llava-next: Improved reasoning, ocr, and world knowledge,” January 2024. [Online]. Available: https://llava-vl.github.io/blog/2024-01-30-llava-next/

    2024

  13. [21]

    Viewpoint integra- tion and registration with vision language foundation model for image change understanding,

    X. Lu, J. Yuan, R. Niu, Y . Hu, and F. Wang, “Viewpoint integra- tion and registration with vision language foundation model for image change understanding,” arXiv preprint arXiv:2309.08585, 2023

    2023 arXiv

  14. [22]

    Oned- iff: A generalist model for image difference,

    E. Hu, L. Guo, T. Yue, Z. Zhao, S. Xue, and J. Liu, “Oned- iff: A generalist model for image difference,” arXiv preprint arXiv:2407.05645, 2024

    2024 arXiv

  15. [23]

    Differential- perceptive and retrieval-augmented mllm for change caption- ing,

    X. Zhang, H. Wen, J. Wu, P. Qin, L. Nie et al. , “Differential- perceptive and retrieval-augmented mllm for change caption- ing,” in ACM International Conference on Multimedia[C], 2024

    2024

  16. [24]

    Cdchat: A large multimodal model for remote sensing change description,

    M. Noman, N. Ahsan, M. Naseer, H. Cholakkal, R. M. An- wer, S. Khan, and F. S. Khan, “Cdchat: A large multimodal model for remote sensing change description,” arXiv preprint arXiv:2409.16261, 2024

    2024 arXiv

  17. [25]

    Changechat: An interactive model for remote sensing change analysis via multimodal instruction tuning,

    P. Deng, W. Zhou, and H. Wu, “Changechat: An interactive model for remote sensing change analysis via multimodal instruction tuning,” arXiv preprint arXiv:2409.08582 , 2024

    2024 arXiv

  18. [26]

    Enhancing perception of key changes in remote sensing image change captioning,

    C. Yang, Z. Li, H. Jiao, Z. Gao, and L. Zhang, “Enhancing perception of key changes in remote sensing image change captioning,” arXiv preprint arXiv:2409.12612 , 2024

    2024 arXiv

  19. [27]

    Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models,

    J. Li, D. Li, S. Savarese, and S. Hoi, “Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models,” arXiv:2301.12597, 2023

    2023 arXiv

  20. [28]

    Gpt-4 technical report,

    OpenAI, “Gpt-4 technical report,” 2023

    2023

  21. [29]

    Hello gpt-4o,

    ——, “Hello gpt-4o,” https://openai.com/index/hello-gpt-4o/, 2024

    2024

  22. [30]

    Gemini: a family of highly capable multimodal models,

    G. Team, R. Anil, S. Borgeaud, Y . Wu, J.-B. Alayrac, J. Yu, R. Soricut, J. Schalkwyk, A. M. Dai, A. Hauth et al., “Gemini: a family of highly capable multimodal models,” arXiv preprint arXiv:2312.11805, 2023

    2023 arXiv

  23. [31]

    Claude-3.5,

    Anthropic, “Claude-3.5,” https://www.anthropic.com/news/ claude-3-5-sonnet, 2024

    2024

  24. [32]

    Flamingo: a visual language model for few-shot learning,

    J.-B. Alayrac, J. Donahue, P. Luc, A. Miech, I. Barr, Y . Has- son, K. Lenc, A. Mensch, K. Millican, M. Reynolds et al. , “Flamingo: a visual language model for few-shot learning,” Advances in Neural Information Processing Systems , vol. 35, pp. 23 716–23 736, 2022

    2022

  25. [33]

    mplug-owl: Modularization empowers large language models with multimodality,

    Q. Ye, H. Xu, G. Xu, J. Ye, M. Yan, Y . Zhou, J. Wang, A. Hu, P. Shi, Y . Shi et al. , “mplug-owl: Modularization empowers large language models with multimodality,” arXiv:2304.14178, 2023

    2023 arXiv

  26. [34]

    Ot- ter: A multi-modal model with in-context instruction tuning,

    B. Li, Y . Zhang, L. Chen, J. Wang, J. Yang, and Z. Liu, “Ot- ter: A multi-modal model with in-context instruction tuning,” arXiv:2305.03726, 2023

    2023 arXiv

  27. [35]

    Visionllm: Large language model is also an open-ended decoder for vision-centric tasks,

    W. Wang, Z. Chen, X. Chen, J. Wu, X. Zhu, G. Zeng, P. Luo, T. Lu, J. Zhou, Y . Qiao et al. , “Visionllm: Large language model is also an open-ended decoder for vision-centric tasks,” arXiv:2305.11175, 2023

    2023 arXiv

  28. [36]

    Kosmos-2: Grounding multimodal large language models to the world,

    Z. Peng, W. Wang, L. Dong, Y . Hao, S. Huang, S. Ma, and F. Wei, “Kosmos-2: Grounding multimodal large language models to the world,” arXiv:2306.14824, 2023

    2023 arXiv

  29. [37]

    Detgpt: Detect what you need via reasoning,

    R. Pi, J. Gao, S. Diao, R. Pan, H. Dong, J. Zhang, L. Yao, J. Han, H. Xu, L. Kong, and T. Zhang, “Detgpt: Detect what you need via reasoning,” arXiv:2305.14167, 2023

    2023 arXiv

  30. [38]

    Deepseek-vl: towards real-world vision-language understanding,

    H. Lu, W. Liu, B. Zhang, B. Wang, K. Dong, B. Liu, J. Sun, T. Ren, Z. Li, Y . Sun et al. , “Deepseek-vl: towards real-world vision-language understanding,” arXiv preprint arXiv:2403.05525, 2024

    2024 arXiv

  31. [39]

    Intern vl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks,

    Z. Chen, J. Wu, W. Wang, W. Su, G. Chen, S. Xing, Z. Muyan, Q. Zhang, X. Zhu, L. Lu, B. Li, P. Luo, T. Lu, Y . Qiao, and J. Dai, “Intern vl: Scaling up vision foundation models and aligning for generic visual-linguistic tasks,” 2024 IEEE/CVF Conference on Computer Vision and P...

    2024

  32. [40]

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

    Z. Chen, W. Wang, H. Tian, S. Ye, Z. Gao, E. Cui, JOURNAL OF LATEX CLASS FILES, VOL. 14, NO. 8, AUGUST 2015 14 W. Tong, K. Hu, J. Luo, Z. Ma, J. Ma, J. Wang, X. wen Dong, H. Yan, H. Guo, C. He, Z. Jin, C. Xu, B. Wang, X. Wei, W. Li, W. Zhang, L. Lu, X. Zhu, T. Lu, D. Lin, and ...

    2015 arXiv

  33. [41]

    Qwen2-vl: Enhancing vision- language model’s perception of the world at any resolution,

    P. Wang, S. Bai, S. Tan, S. Wang, Z. Fan, J. Bai, K. Chen, X. Liu, J. Wang, W. Ge et al. , “Qwen2-vl: Enhancing vision- language model’s perception of the world at any resolution,” arXiv preprint arXiv:2409.12191 , 2024

    2024 arXiv

  34. [42]

    Learning to describe differences between pairs of similar images,

    H. Jhamtani and T. Berg-Kirkpatrick, “Learning to describe differences between pairs of similar images,” in Proceedings of the 2018 Conference on Empirical Methods in Natural Language Processing, 2018, pp. 4024–4034

    2018

  35. [43]

    Robust change cap- tioning,

    D. H. Park, T. Darrell, and A. Rohrbach, “Robust change cap- tioning,” in IEEE/CVF International Conference on Computer Vision[C], 2019, pp. 4623–4632

    2019

  36. [44]

    Viewpoint- agnostic change captioning with cycle consistency,

    H. Kim, J. Kim, H. Lee, H. Park, and G. Kim, “Viewpoint- agnostic change captioning with cycle consistency,” in IEEE/CVF Conference on Computer Vision and Pattern Recog- nition[C], 2021, pp. 2075–2084

    2021

  37. [45]

    Finding it at another side: A viewpoint-adapted matching encoder for change captioning,

    X. Shi, X. Yang, J. Gu, S. R. Joty, and J. Cai, “Finding it at another side: A viewpoint-adapted matching encoder for change captioning,” in European Conference on Computer Vision[C] , 2020, pp. 574–590

    2020

  38. [46]

    Image change captioning by learning from an auxiliary task,

    M. Hosseinzadeh and Y . Wang, “Image change captioning by learning from an auxiliary task,” in IEEE/CVF Conference on Computer Vision and Pattern Recognition[C] , 2021, pp. 2725– 2734

    2021

  39. [47]

    Visual dialog for spotting the differences between pairs of similar images,

    D. Zheng, F. Meng, Q. Si, H. Fan, Z. Xu, J. Zhou, F. Feng, and X. Wang, “Visual dialog for spotting the differences between pairs of similar images,” in ACM International Conference on Multimedia[C], 2022, pp. 5698–5709

    2022

  40. [48]

    Captioning changes in bi-temporal remote sensing images,

    S. Chouaf, G. Hoxha, Y . Smara, and F. Melgani, “Captioning changes in bi-temporal remote sensing images,” in Proc. IEEE Int. Geosci. Remote. Sens. Symp. , 2021, pp. 2891–2894

    2021

  41. [49]

    Change captioning: A new paradigm for multitemporal remote sensing image analysis,

    G. Hoxha, S. Chouaf, F. Melgani, and Y . Smara, “Change captioning: A new paradigm for multitemporal remote sensing image analysis,” IEEE Trans. Geosci. Remote. Sens. , vol. 60, pp. 1–14, 2022

    2022

  42. [50]

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

    S. Chang and P. Ghamisi, “Changes to captions: An attentive network for remote sensing change captioning,” IEEE Transac- tions on Image Processing , vol. 32, pp. 6047–6060, 2023

    2023

  43. [51]

    A deeply supervised attention metric-based network and an open aerial image dataset for remote sensing change detection,

    Q. Shi, M. Liu, S. Li, X. Liu, F. Wang, and L. Zhang, “A deeply supervised attention metric-based network and an open aerial image dataset for remote sensing change detection,” IEEE Transactions on Geoscience and Remote Sensing , pp. 1–16, 2021

    2021

  44. [52]

    A spatial-temporal attention-based method and a new dataset for remote sensing image change detection,

    H. Chen and Z. Shi, “A spatial-temporal attention-based method and a new dataset for remote sensing image change detection,” Remote Sensing , vol. 12, no. 10, p. 1662, 2020

    2020

  45. [53]

    An image is worth 16x16 words: Trans- formers for image recognition at scale,

    A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly et al. , “An image is worth 16x16 words: Trans- formers for image recognition at scale,” arXiv preprint arXiv:2010.11929, 2020

    2010 arXiv

  46. [54]

    Sigmoid loss for language image pre-training,

    X. Zhai, B. Mustafa, A. Kolesnikov, and L. Beyer, “Sigmoid loss for language image pre-training,” 2023

    2023

  47. [55]

    Robust change captioning,

    D. H. Park, T. Darrell, and A. Rohrbach, “Robust change captioning,” in Proceedings of the IEEE/CVF International Conference on Computer Vision , 2019, pp. 4624–4633

    2019

  48. [56]

    Expressing visual relationships via language,

    H. Tan, F. Dernoncourt, Z. Lin, T. Bui, and M. Bansal, “Expressing visual relationships via language,” arXiv preprint arXiv:1906.07689, 2019

    1906 arXiv

  49. [57]

    Learning to describe differences between pairs of similar images,

    H. Jhamtani and T. Berg-Kirkpatrick, “Learning to describe differences between pairs of similar images,” arXiv preprint arXiv:1808.10584, 2018

    2018 arXiv

  50. [58]

    The stvchrono dataset: Towards continuous change recognition in time,

    Y . Sun, Y . Qiu, M. Khan, F. Matsuzawa, and K. Iwata, “The stvchrono dataset: Towards continuous change recognition in time,” in Proceedings of the IEEE/CVF Conference on Com- puter Vision and Pattern Recognition , 2024, pp. 14 111–14 120

    2024

  51. [59]

    Vis- min: Visual minimal-change understanding,

    R. Awal, S. Ahmadi, L. Zhang, and A. Agrawal, “Vis- min: Visual minimal-change understanding,” arXiv preprint arXiv:2407.16772, 2024

    2024 arXiv

  52. [60]

    Changesim: Towards end-to-end online scene change detection in industrial indoor environments,

    J.-M. Park, J.-h. Jang, S.-M. Yoo, S.-K. Lee, U.-h. Kim, and J.-H. Kim, “Changesim: Towards end-to-end online scene change detection in industrial indoor environments,” in 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) . IEEE, 2021. [Online]. A...

    2021 arXiv

  53. [61]

    Semantic change detection with asymmetric siamese networks,

    K. Yang, G.-S. Xia, Z. Liu, B. Du, W. Yang, M. Pelillo, and L. Zhang, “Semantic change detection with asymmetric siamese networks,” 2020

    2020

  54. [62]

    Robust change captioning,

    D. H. Park, T. Darrell, and A. Rohrbach, “Robust change captioning,” in 2019 IEEE/CVF International Conference on Computer Vision (ICCV) , 2019, pp. 4623–4632

    2019

  55. [63]

    Describing and localizing multiple changes with transformers,

    Y . Qiu, S. Yamamoto, K. Nakashima, R. Suzuki, K. Iwata, H. Kataoka, and Y . Satoh, “Describing and localizing multiple changes with transformers,” in 2021 IEEE/CVF International Conference on Computer Vision (ICCV) , 2021, pp. 1951–1960

    2021

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.