LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents

Alexandra Fortacz-Lazan; Franziska H\"ubl; Ivan Majic; Krzysztof Janowicz; Meilin Shi; Mina Karimi; Zexian Huang; Zilong Liu

REVIEW 4 major objections 6 minor 14 references

LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents

T0 review · 4 major / 6 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read Recent OpenAI models still lose substantial spatial information when transferring an image of a color grid to text and back to an image, with errors growing sharply with grid size and color count.

desk verdict A useful, cheap benchmark for cross-modal spatial transfer that shows a real failure trend, but the quantitative error rates rest on a manual, unreproducible evaluation step. read the letter →

arxiv 2608.06948 v1 pith:J6MW3NRB submitted 2026-08-07 cs.AI

Ivan Majic , Zexian Huang , Franziska Hübl , Krzysztof Janowicz , Meilin Shi , Mina Karimi , Zilong Liu , Alexandra Fortacz-Lazan This is my paper · ORCID

classification cs.AI

keywords LargeMultimodalModelModalityTransferSpatialReasoningGISAgentGeoAIHallucinationEarthMover'sDistanceLevenshtein

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 proposes a modality transfer task that sends an image of an N×N grid of colored squares through an image-to-text prompt and then a text-to-image prompt, and measures how much spatial information survives the round trip. The authors claim that recent OpenAI models (gpt-5 for reading the image, gpt-image-1 for redrawing it) still struggle with this transfer, even on the simplest 5×5 three-color grids. They find that errors rise abruptly as the grid grows beyond 7×7 and as more colors are added, with 5-color configurations often collapsing into severe hallucinations. The paper argues that near-lossless modality transfer is a prerequisite for autonomous GIS agents, which must move seamlessly between maps and textual descriptions in real workflows.

What carries the argument

The central object is the modality transfer task itself, a two-stage pipeline: Prompt 1 asks gpt-5 to explain an input grid image in text usable by other LMMs, and Prompt 2 hands that text to gpt-image-1 to redraw the scene. Evaluation converts ground truth, generated text, and generated images into discrete color matrices, then applies two complementary metrics: a 1D Levenshtein distance on the flattened matrices for sequence/topology errors, and a class-wise 2D Earth Mover's Distance (solved via optimal transport) for geometric displacement, with a maximum-diagonal penalty for missing or hallucinated cells. A CIEDE2000 perceptual threshold maps sampled output colors to the five palette classes or flags them as out-of-bounds hallucinations.

What would settle it

Bypass the image-to-text stage and feed the ground-truth color matrix directly to gpt-image-1: if the regenerated images match the originals with near-zero Levenshtein and EMD, the bottleneck is in image-to-text; if errors remain large, the bottleneck is in text-to-image. Independently, replace the manual GUI color sampling with an automated computer-vision extraction and re-run the 900-image pipeline to check whether the reported error magnitudes and failure thresholds persist.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that lossless modality transfer between image and text remains profoundly difficult for state-of-the-art LMMs. Using 900 procedurally generated ground-truth grids, the authors show that the image-to-text stage (gpt-5) already introduces sequence errors that grow with grid size, and the text-to-image stage (gpt-image-1) amplifies them, producing images that deviate from the original by large Levenshtein distances and Earth Mover's Distances. The failure taxonomy places most low-complexity cases in 'minor attribute error', but with a 5-color palette the outputs cluster in the 'severe hallucination' quadrant, indicating a complete loss of both topology and geometry. The authors conclude that strong multi-modal alignment, not just scale or prior knowledge, is required for reliable geospatial understanding.

Load-bearing premise

The quantitative results depend on a human operator manually sampling the cell colors of each generated image and on the authors' inclusion/exclusion judgments for distorted outputs, a step the paper concedes is vulnerable to human error and subjective interpretation.

Editorial extensions

If this is right

Autonomous GIS agents that rely on LMMs for cross-modal steps will inherit these spatial losses unless the underlying models are explicitly aligned for modality transfer.
The proposed task can serve as a benchmark for tracking progress in multimodal spatial understanding across model families and versions.
The abrupt degradation at larger grids and color counts indicates a capacity bottleneck that prompt engineering alone is unlikely to remove.
Diagnosing errors separately in image-to-text and text-to-image directions becomes necessary to pinpoint where spatial information is lost.

Reading between the lines

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

The paper tests only two proprietary zero-shot models; we infer that running the same benchmark on open-weights models and with few-shot prompting would clarify whether the failure is intrinsic to multimodal alignment or partly an artifact of prompt style.
Because the task uses uniformly colored abstract grids, we infer that real-world map images with contextual cues (road networks, labels, textures) could actually improve transfer performance, as the paper itself hints; this is testable by swapping grid images for real land-cover map tiles.
The manual GUI sampling step is a likely source of noise in the reported metrics; we infer that an automated computer-vision parser could change the absolute error values while preserving the qualitative trend, and would make the benchmark reproducible at scale.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

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

Desk Editor's Note

read the letter

The genuinely new thing here is the closed-loop task: image to text to image on colored grids, with the text passed between two different LMM instances. Prior work looked at spatial reasoning in one modality or the other, not at lossless transfer across both. The paper also ships code, input images, and prompt outputs, so the benchmark is cheap to reuse and extend. The qualitative failure trend is credible: the figures show the models losing structure as grid size and color count go up, and the authors are appropriately careful in their limitations section.

The soft spots are real and they hit the quantitative claims. The stress-test note is right about the Section 5.2 contradiction: one paragraph says invalid outputs were assigned maximum errors, the next says they were omitted from quantitative analysis. Those are opposite treatments, and the reported averages change meaning depending on which actually happened. The manual GUI sampling of generated images is a second problem; the authors admit it is vulnerable to human error and subjective judgment. And the CIEDE2000 threshold tau is never given a value, so no one can reproduce the mapping from pixels to categorical colors. On top of that, the text-to-image stage is not isolated from the image-to-text stage, so the reported error rates conflate failures in the two directions.

That said, the central observation survives these problems. The qualitative examples independently show catastrophic failures at higher complexity, and the paper does not overclaim beyond its evidence. The issues make the specific numeric error rates insecure, but they do not destroy the conclusion that LMMs struggle with this task. The contradiction and the missing tau are fixable in revision; the manual sampling could be replaced or at least supplemented with an automated method.

Who is this for? GeoAI and spatial-reasoning researchers who want a simple diagnostic for multimodal integration. It will likely be useful as a benchmark even if the current evaluation needs hardening. I would not cite the numbers as ground truth yet, but I would send the paper to peer review with a clear request to resolve the invalid-output contradiction, report tau, and add an automated variant of the sampling step. The task definition is a contribution worth keeping.

Referee Report

4 major / 6 minor

Summary. The paper proposes a “modality transfer task” in which a procedurally generated image of an N×N grid of colored squares is first described by an LMM (OpenAI gpt-5) and then re-generated by an image-generation LMM (OpenAI gpt-image-1). All representations are converted into categorical color matrices, and the authors compute 1D Levenshtein distance and a class-wise spatial Earth Mover’s Distance, using CIEDE2000 color thresholding for generated images. Across 900 trials with grid sizes N = 5 to 10 and palettes of 3 to 5 colors, the paper reports increasing spatial loss with complexity, an abrupt capability drop at N = 7 for the 3-color case, and concludes that lossless image-to-text-to-image transfer remains difficult for current LMMs and is a prerequisite for autonomous GIS agents.

Significance. The task is simple, visually intuitive, and well matched to the GIS/land-cover motivation. Strengths include the public repository, the large number of trials, the use of a perceptually motivated color-difference formula, and a diagnostic separation between sequential and geometric error. If the measurements were reproducible, the observed failure trend would be a meaningful caution for agentic GIS workflows. However, the quantitative results currently depend on a manual GUI-based color-sampling step, an unreported threshold parameter, and contradictory inclusion/exclusion rules for invalid outputs. The qualitative examples in Figure 11 support the general claim that current models make substantial modality-transfer errors, but the specific error magnitudes and the claimed abrupt drop at N = 7 are not yet established by a secure evaluation pipeline.

major comments (4)

[Section 5.2] The handling of invalid generated images is internally contradictory. The text states that cases with unsampleable colors “were assigned maximum errors in the evaluation,” and then later states that severe generative failures “had to be omitted from the quantitative analysis.” These two rules cannot both hold. If invalid outputs are omitted, every reported average is conditional on the surviving samples and understates the true failure rate; if they are assigned maximum errors, the magnitude of the trends in Figures 7–9 is partly determined by the inclusion rule rather than by measured grid content. Because the central claim of an abrupt capability drop at N = 7 is based on these aggregates, the authors must specify one reproducible rule, apply it uniformly, and report results under both treatments or provide a sensitivity analysis.
[Section 3.3 and Section 5.4] The Levenshtein and EMD values are computed from matrices extracted by a human clicking cell centers in a GUI. The authors themselves acknowledge that this step is “vulnerable to human errors and subjective interpretation of the operator.” No inter-rater reliability, automated verification, or release of the exact manually sampled matrices is provided. Without an objective or validated extraction protocol, the reported error values are not reproducible. Please either automate the extraction, provide a second-annotator agreement analysis on a subsample, or release the raw output images together with the exact sampled matrices and a detailed decision protocol.
[Section 3.4] The CIEDE2000 threshold τ is a free parameter in the color-assignment rule, but no numeric value is ever reported. Generated colors with minimum ΔE00 ≤ τ are mapped to a palette class, while colors exceeding τ are marked as hallucinated “h.” Both the Levenshtein and the EMD measures therefore depend directly on τ, and the reported values cannot be reproduced without it. Please report the exact τ used, and ideally show the sensitivity of the main scaling trends to reasonable variations of τ.
[Section 5.4] The evaluation is joint: the text-to-image stage always receives the potentially erroneous output of the image-to-text stage. The authors note that it would be informative to use ground-truth text for Prompt 2, but they do not provide such an ablation. As a result, the diagnostic claims about “cascading” failures and about the relative contribution of Prompt 1 versus Prompt 2 cannot be separated from the end-to-end pipeline. Please add the ground-truth-text condition, or clearly restrict all conclusions to the end-to-end pipeline rather than attributing failures to a specific stage.

minor comments (6)

[Section 4] The references to “Figure 8b” and “Figure 8c” for the 4- and 5-color palette results appear to be mislabeled; the corresponding subfigures are Figure 7b and Figure 7c.
[Figure 9 caption] The caption contains a typo: “Eorror” should be “Error.”
[Section 5.2 and Section 5.4] There are several typos: “ommitted” should be “omitted,” “wholistic” should be “holistic,” and “zero-short” should be “zero-shot.”
[Listing 1] The prompt text contains spacing artifacts such as “re pr es en t” and “re as on in g”; the prompt should be typeset as verbatim text so it is readable.
[Figure 8] The aggregated plots would benefit from error bars or per-condition distributions; with 50 trials per condition, the stability of the “abrupt drop” at N = 7 is currently not quantified.
[Table 1] The diagnostic categories in Table 1 use “Low” and “High” without operational thresholds; Figure 9 shows quadrant guides, but the cutoff values are never defined, making the category labels hard to interpret quantitatively.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the paper reports an empirical measurement with independent ground-truth and model-generated outputs, and its self-citations are not load-bearing.

full rationale

The central claim is an empirical result: OpenAI LMMs lose spatial information when performing image-to-text-to-image transfer on procedurally generated colored grids. The ground-truth matrices are generated independently of the model outputs, and the evaluation metrics (Levenshtein distance and spatial EMD) compare those independent representations. No fitted parameter is renamed as a prediction, no equation defines the output in terms of the input by construction, and no uniqueness theorem is imported from the authors' prior work. The self-citations appear only in framing and related-work contexts (e.g., the GeoAI agenda and prior spatial-reasoning studies) and are not the evidence supporting the reported failure rates. The manual GUI-based color sampling and the inclusion/exclusion decisions for malformed outputs are reproducibility and validity concerns, not circularity: they affect whether the measurements are accurate, but they do not make the measured result equivalent to its own assumptions. The unreported threshold tau is a transparency issue, not a self-referential derivation. Accordingly, the paper's derivation chain is self-contained and no circular step can be exhibited.

Assumptions & free parameters 1 free parameters · 5 assumptions · 0 invented entities

No model was fitted to data, so the ledger is small. The main non-standard inputs are the unspecified color threshold tau, the manual sampling premise, and the framing assumption that lossless modality transfer is required for GIS agents.

free parameters (1)

CIEDE2000 threshold tau = not stated in paper
Section 3.4 defines tau as the perceptual distance below which generated colors map to palette colors; the numeric value is not given in the manuscript, so all color classification results depend on an unspecified hand-set constant.

assumptions (5)

domain assumption Lossless modality transfer is a prerequisite for autonomous GIS agents
Stated in the introduction and conclusion as the motivation; not derived or tested, and it frames the paper's significance.
domain assumption Manual GUI sampling accurately converts generated images into discrete color matrices
Section 3.3; all image-side metrics depend on this subjective extraction step.
domain assumption Invalid generated images can be excluded or assigned maximum error without biasing aggregate results
Section 5.2 contains both 'omitted completely' and 'assigned maximum errors'; the choice changes error magnitudes and failure categorization.
domain assumption The OpenAI models tested are representative enough to support statements about state-of-the-art LMMs
Section 5.4 limits the study to gpt-5 and gpt-image-1 and only speculates that other models would also fail.
standard math Levenshtein distance and the assignment-based EMD quantify spatial information loss
Section 3.4; these are established metrics, though flattening the grid discards some 2D structure.

how reviews work

0 comments

Cite this review

Pith. "Pith review of LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents." pith.science (2026). https://pith.science/paper/J6MW3NRB

@misc{pith2026260806948,
  author       = {Pith},
  title        = {Pith review of: LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/J6MW3NRB}},
  note         = {Machine review of arXiv:2608.06948}
}

read the original abstract

AI models are becoming increasingly adept at understanding and processing spatial information, thereby facilitating agentic problem-solving in spatial tasks and workflows. However, most of the research on their spatial capabilities (e.g., spatial reasoning) has focused on the textual modality as input and output. This contrasts with the human approach to GIS workflows, where text and visual modalities are often used together, interchangeably, and in a complementary manner. Thus, to truly achieve an automated GIS analysis pipeline or carry out human-designed GIS workflows, AI models --- Large Multimodal Models (LMMs) in particular --- need to be able to seamlessly transition between image- and text-based modalities that are traditionally used in such workflows. We present a modality transfer task that (1) asks an LMM to first describe an input image of colored squares in a regular grid, and (2) asks a new LMM instance to re-generate an image of the original spatial scene using the textual description output by the former model. This task quantifies the ability of LMMs to transfer spatial information between image and text modalities. Ultimately, by examining the modality transfer capability of LMMs through the lens of spatial information theory, this work highlights a critical bottleneck: achieving strong and robust geospatial understanding in LMMs requires rigorous, multi-modal alignment. Our results indicate that recent LMMs (here from OpenAI) still struggle with modality transfer, when tasked with re-generating an image of a simple spatial grid of color squares.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

14 extracted references · 7 canonical work pages

[2]

Can large language models reason about the region connection calculus?arXiv preprint arXiv:2411.19589,

4 Anthony G Cohn and Robert E Blackwell. Can large language models reason about the region connection calculus?arXiv preprint arXiv:2411.19589,

arXiv
[3]

Evaluating the Ability of Large Language Models to Reason About Cardinal Directions

5 Anthony G Cohn and Robert E Blackwell. Evaluating the Ability of Large Language Models to Reason About Cardinal Directions. In Benjamin Adams, Amy L. Griffin, Simon Scheider, and GrantMcKenzie, editors,16th International Conference on Spatial Information Theory (COSIT 2024), volume 315 ofLeibniz International Proceedings in Informatics (LIPIcs), pages 2...

work page 2024
[4]

URL:https: //drops.dagstuhl.de/entities/document/10.4230/LIPIcs.COSIT.2024.28, doi:10.4230/ LIPIcs.COSIT.2024.28

Schloss Dagstuhl – Leibniz-Zentrum für Informatik. URL:https: //drops.dagstuhl.de/entities/document/10.4230/LIPIcs.COSIT.2024.28, doi:10.4230/ LIPIcs.COSIT.2024.28. 6 Eirinaios-Odysseas Gardelakos, Vasileios Kyriakopoulos, Despina-Athanasia Pantazi, Orestis- Minas Kapopoulos, Maria Tsourma, and Manolis Koubarakis. Can Large Reasoning Models Reason about S...

doi:10.4230/lipics.cosit.2024.28 2024
[8]

URL: /doi/pdf/10.1080/19475683.2025.2552161?download=true, doi:10.1080/19475683.2025.2552161. 11 M. R. Luo, G. Cui, and B. Rigg. The development of the CIE 2000 colour- difference formula: CIEDE2000.Color Research and Application, 26:340–350, 10

arXiv 2025
[12]

mdpi.com/2413-8851/3/2/48/htmhttps://www.mdpi.com/2413-8851/3/2/48, doi:10.3390/ urbansci3020048

URL: https://www. mdpi.com/2413-8851/3/2/48/htmhttps://www.mdpi.com/2413-8851/3/2/48, doi:10.3390/ urbansci3020048. 16 Ivan Majic, Zhangyu Wang, Krzysztof Janowicz, and Mina Karimi. Spatial Task-Explicity Mat- ters in Prompting Large Multimodal Models for Spatial Planning.GeoAI 2024 - Proceedings of the 7th ACM SIGSPATIAL International Workshop on AI for ...

work page
[13]

17 Manuel Melgosa

URL:https://dl.acm.org/doi/pdf/10.1145/3687123.3698293, doi:10.1145/3687123.3698293. 17 Manuel Melgosa. Testing CIELAB-based Color-Difference Formulas.Color Research & Application, 25(1):49–55, 2000.doi:10.1002/(SICI)1520-6378(200002)25:1<49::AID-COL7> 3.0.CO;2-4. 18 Yossi Rubner, Carlo Tomasi, and Leonidas J Guibas. The earth mover’s distance as a metric...

arXiv
[1965]

org/CorpusID:60827152

URL: https://api.semanticscholar. org/CorpusID:60827152. COSIT 2026 14:20 LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents 10 Zhenlong Li, Huan Ning, Song Gao, Krzysztof Janowicz, Wenwen Li, Samantha T. Arundel, Chaowei Yang, Budhendra Bhaduri, Shaowen Wang, A. Xing Zhu, Mark Gahegan, Shashi Shekhar, Xinyue Ye, Grant McKenzie, Guido Cervon...

work page 2026
[2001]

1002/col.1049https://onlinelibrary.wiley.com/doi/10.1002/col.1049, doi:10.1002/ col.1049

URL: /doi/pdf/10.1002/col.1049https://onlinelibrary.wiley.com/doi/abs/10. 1002/col.1049https://onlinelibrary.wiley.com/doi/10.1002/col.1049, doi:10.1002/ col.1049. 12 Gengchen Mai, Weiming Huang, Jin Sun, Suhang Song, Deepak Mishra, Ninghao Liu, Song Gao, Tianming Liu, Gao Cong, Yingjie Hu, et al. On the opportunities and challenges of foundation models f...

arXiv

Show all 14 references

[2015]

Evaluating Intrinsic Geospatial Topological Reasoning in LLMs

20 Shaolin Xie, Shang-Ling Hsu, Qihan Zhang, Yiming Gao, Cyrus Shahabi, and Ibrahim Sabek. Evaluating Intrinsic Geospatial Topological Reasoning in LLMs. InProceedings of the 1st ACM SIGSPATIAL International Workshop on Generative and Agentic AI for Multi-Modality Space-Time I...

2025
[2019]

14 Ivan Majic, Elham Naghizade, Stephan Winter, and Martin Tomko

URL:https://josis.org/index.php/josis/article/ view/99,doi:10.5311/JOSIS.2019.18.459. 14 Ivan Majic, Elham Naghizade, Stephan Winter, and Martin Tomko. There is no way! Ternary qualitative spatial reasoning for error detection in map data.Transactions in GIS, 25:2048– 2073, 8

2019 doi
[2020]

8 Yuhan Ji, Song Gao, Ying Nie, Ivan Majić, and Krzysztof Janowicz

URL: https://www.tandfonline.com/doi/pdf/10.1080/13658816.2019.1684500, doi:10.1080/13658816.2019.1684500. 8 Yuhan Ji, Song Gao, Ying Nie, Ivan Majić, and Krzysztof Janowicz. Foundation models for geospatial reasoning: assessing the capabilities of large language models in und...

2019
[2021]

15 Ivan Majic and Elek Pafka

URL:/doi/pdf/10.1111/tgis.12765https://onlinelibrary.wiley.com/doi/ abs/10.1111/tgis.12765https://onlinelibrary.wiley.com/doi/10.1111/tgis.12765, doi:10.1111/tgis.12765. 15 Ivan Majic and Elek Pafka. AwaP-IC—An Open-Source GIS Tool for Measuring Walk- able Access.Urban Science...

doi
[2024]

Arc prize 2024: Technical report.arXiv preprint arXiv:2412.04604,

3 Francois Chollet, Mike Knoop, Gregory Kamradt, and Bryan Landers. Arc prize 2024: Technical report.arXiv preprint arXiv:2412.04604,

2024 arXiv
[2025]

7 Krzysztof Janowicz, Song Gao, Grant McKenzie, Yingjie Hu, and Budhendra Bhaduri

Association for Computing Machinery.doi:10.1145/3764912.3770841. 7 Krzysztof Janowicz, Song Gao, Grant McKenzie, Yingjie Hu, and Budhendra Bhaduri. GeoAI: spatially explicit artificial intelligence techniques for geographic knowledge discov- ery and beyond.International Journa...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.

[1] [2]

Can large language models reason about the region connection calculus?arXiv preprint arXiv:2411.19589,

4 Anthony G Cohn and Robert E Blackwell. Can large language models reason about the region connection calculus?arXiv preprint arXiv:2411.19589,

arXiv

[2] [3]

Evaluating the Ability of Large Language Models to Reason About Cardinal Directions

5 Anthony G Cohn and Robert E Blackwell. Evaluating the Ability of Large Language Models to Reason About Cardinal Directions. In Benjamin Adams, Amy L. Griffin, Simon Scheider, and GrantMcKenzie, editors,16th International Conference on Spatial Information Theory (COSIT 2024), volume 315 ofLeibniz International Proceedings in Informatics (LIPIcs), pages 2...

work page 2024

[3] [4]

URL:https: //drops.dagstuhl.de/entities/document/10.4230/LIPIcs.COSIT.2024.28, doi:10.4230/ LIPIcs.COSIT.2024.28

Schloss Dagstuhl – Leibniz-Zentrum für Informatik. URL:https: //drops.dagstuhl.de/entities/document/10.4230/LIPIcs.COSIT.2024.28, doi:10.4230/ LIPIcs.COSIT.2024.28. 6 Eirinaios-Odysseas Gardelakos, Vasileios Kyriakopoulos, Despina-Athanasia Pantazi, Orestis- Minas Kapopoulos, Maria Tsourma, and Manolis Koubarakis. Can Large Reasoning Models Reason about S...

doi:10.4230/lipics.cosit.2024.28 2024

[4] [8]

URL: /doi/pdf/10.1080/19475683.2025.2552161?download=true, doi:10.1080/19475683.2025.2552161. 11 M. R. Luo, G. Cui, and B. Rigg. The development of the CIE 2000 colour- difference formula: CIEDE2000.Color Research and Application, 26:340–350, 10

arXiv 2025

[5] [12]

mdpi.com/2413-8851/3/2/48/htmhttps://www.mdpi.com/2413-8851/3/2/48, doi:10.3390/ urbansci3020048

URL: https://www. mdpi.com/2413-8851/3/2/48/htmhttps://www.mdpi.com/2413-8851/3/2/48, doi:10.3390/ urbansci3020048. 16 Ivan Majic, Zhangyu Wang, Krzysztof Janowicz, and Mina Karimi. Spatial Task-Explicity Mat- ters in Prompting Large Multimodal Models for Spatial Planning.GeoAI 2024 - Proceedings of the 7th ACM SIGSPATIAL International Workshop on AI for ...

work page

[6] [13]

17 Manuel Melgosa

URL:https://dl.acm.org/doi/pdf/10.1145/3687123.3698293, doi:10.1145/3687123.3698293. 17 Manuel Melgosa. Testing CIELAB-based Color-Difference Formulas.Color Research & Application, 25(1):49–55, 2000.doi:10.1002/(SICI)1520-6378(200002)25:1<49::AID-COL7> 3.0.CO;2-4. 18 Yossi Rubner, Carlo Tomasi, and Leonidas J Guibas. The earth mover’s distance as a metric...

arXiv

[7] [1965]

org/CorpusID:60827152

URL: https://api.semanticscholar. org/CorpusID:60827152. COSIT 2026 14:20 LMM Modality Transfer: A Pre-requisite for Autonomous GIS Agents 10 Zhenlong Li, Huan Ning, Song Gao, Krzysztof Janowicz, Wenwen Li, Samantha T. Arundel, Chaowei Yang, Budhendra Bhaduri, Shaowen Wang, A. Xing Zhu, Mark Gahegan, Shashi Shekhar, Xinyue Ye, Grant McKenzie, Guido Cervon...

work page 2026

[8] [2001]

1002/col.1049https://onlinelibrary.wiley.com/doi/10.1002/col.1049, doi:10.1002/ col.1049

URL: /doi/pdf/10.1002/col.1049https://onlinelibrary.wiley.com/doi/abs/10. 1002/col.1049https://onlinelibrary.wiley.com/doi/10.1002/col.1049, doi:10.1002/ col.1049. 12 Gengchen Mai, Weiming Huang, Jin Sun, Suhang Song, Deepak Mishra, Ninghao Liu, Song Gao, Tianming Liu, Gao Cong, Yingjie Hu, et al. On the opportunities and challenges of foundation models f...

arXiv