Pith. sign in

REVIEW 4 major objections 5 minor 40 references

Multimodal graph representation learning for website generation based on visual sketch

T0 review · 4 major / 5 minor · reviewed 2026-08-16 · 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 that feeding a vision-language model an explicit graph of webpage components — text blocks from OCR and visual regions from segmentation, linked by spatial overlap — improves the HTML it generates from a screenshot.

desk verdict The graph-versus-no-graph ablation is confounded by added per-component CLIP features, so the central claim isn't established—but the pipeline is a reasonable incremental contribution worth a conditional referee. read the letter →

arxiv 2504.18729 v1 pith:DZVPSMPG submitted 2025-04-25 cs.LG

Tung D. Vu , Chung Hoang , Truong-Son Hy This is my paper · ORCID
classification cs.LG
keywords Design2Codemultimodalgraphvision-languagemodelneuralnetworkHTMLcodegenerationwebpagelayoutOCRandsegmentationUI-to-code
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 vision-language model generating HTML from a webpage screenshot benefits from being given an explicit multimodal graph of the page's components. The graph is built by extracting text with OCR, segmenting the remaining visual regions, and linking the components so that edges encode semantic and spatial relationships. The authors train a graph-enhanced VLM in which a graph convolutional network encodes this structure and interleaved gated cross-attention layers inject it into a frozen pretrained language model. On a synthetic benchmark they report consistent gains over the same model without the graph, especially on layout metrics such as block matching and position. If the claim holds, explicit structural conditioning offers a practical route to improving layout fidelity in design-to-code systems without scaling model size.

What carries the argument

The machinery is the multimodal component graph and the gated cross-attention that fuses it into the language model. Each node is one extracted webpage component, with textual nodes coming from OCR and visual nodes from segmentation, and each edge encodes either full semantic connectivity among text nodes or spatial overlap above 80 percent IoU between visual-visual and text-visual nodes. A graph convolutional network, seeded with CLIP embeddings of the components, produces node representations that are passed alongside whole-screenshot features, reduced to a fixed number of tokens by a Perceiver resampler, into gated cross-attention layers interleaved between pretrained LM blocks. This design is what carries the argument: it makes page structure explicit and directly attendable at every generation step rather than leaving layout implicit in the screenshot.

What would settle it

Measure precision and recall of the extracted component graph against ground-truth layout annotations (or the DOM tree) on the WebSight test set, then compare OURS-graph fed the extracted graph against the same model fed a ground-truth graph. If the ground-truth graph does not outperform the extracted one, or if randomly corrupted graphs do not degrade performance, then the claim that explicit graph structure drives the reported gains is contradicted.

Watch

Extended reading notes

Core claim

The central claim is that the multimodal component graph is the active ingredient: adding it to a VLM improves HTML code generation on the metrics that measure visual layout, and does so beyond what the screenshot alone provides. The graph is constructed by extracting textual nodes with OCR, masking the text, segmenting the remaining visual nodes, and connecting text nodes to one another while linking visual and text-visual nodes by bounding-box overlap above 80 percent IoU. A graph convolutional network seeded with CLIP component embeddings encodes the graph, and these embeddings are fused into the language model through gated cross-attention interleaved with pretrained layers. In the reported experiments the graph-equipped model beats its no-graph counterpart on the WebSight benchmark across Block-Match, Text, Position, Color, and CLIP metrics, while on the harder Design2Code benchmark the comparison is mixed, with gains on some layout metrics offset by small drops on text and color.

Load-bearing premise

The load-bearing premise is that the OCR-plus-segmentation pipeline extracts webpage components faithfully enough that the graph passed to the model reflects the page's true structure; the paper never measures OCR or segmentation error and does not ablate the extraction step.

Editorial extensions

If this is right

  • On the WebSight benchmark, graph conditioning improves every reported metric over the no-graph model, with the largest gains in Block-Match (24.94 vs 21.60) and Position (70.52 vs 66.21).
  • The graph-enhanced model also improves BLEU, HTML-BLEU, SSIM, and TreeBLEU on the same benchmark, indicating that code and structure fidelity improve together with visual fidelity.
  • The text-mask-then-segment pipeline is presented as the reason non-text components are extracted cleanly, making the two-step extraction a load-bearing part of the method.
  • Because the graph is fused through interleaved cross-attention, the approach can be layered onto a frozen pretrained language model rather than training a new model from scratch.
  • On the real-world Design2Code benchmark, the graph's advantage is smaller and inconsistent, suggesting the benefit is most reliable on synthetic pages with regular layouts.

Reading between the lines

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

  • A natural test the paper does not run is to vary the edge rule: replacing the fixed 80 percent IoU threshold with containment, alignment, or learned edges would show how much of the gain comes from the edge definition itself.
  • The graph's value is likely coupled to the quality of the OCR and segmentation pipeline; deliberately corrupting the graph, by shuffling nodes or deleting edges, and measuring the performance drop would isolate the structural contribution from the visual one.
  • The same conditioning pattern may transfer to other structured generation tasks, such as mobile UI code or document layout generation, wherever components and spatial relations can be extracted from an input image.
  • The mixed Design2Code results suggest that on real pages graph-induced structure may help some aspects while extraction errors hurt others, so an extension that falls back to vision-only when extraction confidence is low is plausible.
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

4 major / 5 minor

Summary. The paper proposes a graph-enhanced vision-language model for the Design2Code task, converting webpage screenshots into HTML code. The pipeline extracts textual components with PaddleOCR and masks them, then segments remaining visual components with SAM; these components become nodes in a multimodal graph with three edge types (text-text fully connected, visual-visual and text-visual by IoU > 0.8). Node features are CLIP embeddings refined by a GCN, and the resulting graph embedding conditions a frozen language model via gated cross-attention layers alongside the full-page image. The authors train on a 20,000-sample subset of WebSight v0.2 and evaluate on a 500-sample WebSight subset and the Design2Code benchmark, comparing OURS-graph against OURS-no-graph, Gemini prompting, and three open VLMs using visual and code metrics.

Significance. If the central claim were established, the contribution would be useful: explicit component graphs are a sensible inductive bias for layout-aware HTML generation, and the paper provides a clear pipeline, a code link, and an honest limitation statement. However, the current evidence does not isolate the effect of graph structure, because the OURS-graph variant differs from OURS-no-graph in both graph edges and per-component CLIP features. In addition, the abstract's efficiency claim is never measured, and the Design2Code results are mixed or reversed on several content metrics. The paper is a reasonable systems proposal that needs a tighter experimental design before its main claims can be accepted.

major comments (4)
  1. [Section 3.2 and Tables 2–3] The headline ablation is confounded: OURS-graph initializes each GCN node with a CLIP embedding of the corresponding OCR/SAM component, whereas OURS-no-graph, as described, appears to receive only the full-page image through the Vision Encoder. The WebSight gains (e.g., Block-Match 24.94 vs 21.60, Color 75.41 vs 69.70) can therefore be attributed to the added per-component CLIP features rather than to graph message passing. A control that keeps the component features but removes the edges (or randomly rewires them) is needed to support the paper's attribution of improvements to multimodal graph representation learning.
  2. [Section 4.3 and Tables 2–3] The text in Section 4.3 states that on the Design2Code benchmark 'OURS-graph maintains superiority over OURS-no-graph', but Table 2 shows the opposite for Text (47.15 vs 50.15), Color (40.07 vs 43.30), and CLIP (82.63 vs 82.90), and Table 3 shows TreeBLEU slightly lower (16.59 vs 16.71). This overstatement, together with the abstract's claim of 'significant improvements in both accuracy and efficiency', is not supported: no efficiency or runtime measurements appear anywhere in the paper.
  3. [Section 4.1 and Section 3] The experimental section lacks essential details: no training hyperparameters, model size, number of steps, batch size, learning rate, or number of seeds are reported, and Tables 2–3 show no error bars. Given that many reported differences are small (e.g., Design2Code CLIP 82.63 vs 82.90), it is not possible to assess whether the observed gains are statistically meaningful. Furthermore, the component-extraction pipeline is a critical assumption that is never validated: the paper does not report OCR or SAM accuracy on the test sets, nor does it ablate the impact of extraction errors on the generated HTML.
  4. [Section 3.1 and Section 4.2] The graph construction uses two ad-hoc heuristics without sensitivity analysis: all textual nodes are fully connected, and a fixed IoU threshold of 0.8 defines visual-visual and text-visual edges. These choices are free parameters of the method, and the paper provides no experiments varying them or justifying them beyond intuition. Since the central claim depends on the graph structure, the robustness of the results to these thresholds should be demonstrated.
minor comments (5)
  1. [Section 4.3 and Table 3 caption] The caption of Table 3 contains a typo, 'TrueBLEU', which should read 'TreeBLEU'.
  2. [Throughout] The benchmark name is inconsistently capitalized as 'Websight' in some places and 'WebSight' in others; the authors should standardize this.
  3. [Section 3.2] The heading 'Vision Encoder with Perceiver Resampler' is duplicated in the text; please remove the repeated phrase.
  4. [References] Several references have broken or awkwardly line-broken URLs (e.g., the OpenFlamingo and UNITER entries); please format them consistently with the journal's style.
  5. [Section 4.2] The baselines are described as 'state-of-the-art open-sourced VLMs', but they are used in a zero-shot setting with no fine-tuning, while OURS is fine-tuned; this comparison should be framed more carefully to avoid implying fairness that is not established.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's central claim is an empirical comparison, and its graph construction is not definitionally tied to the evaluated outputs.

full rationale

The paper's central contribution is an experimental architecture comparison (OURS-graph vs OURS-no-graph vs prompting baselines) on WebSight and Design2Code benchmarks. No equation in the paper defines the predicted HTML metrics in terms of the graph inputs: the graph is built from OCR and SAM component boxes derived from the input screenshot, while the reported metrics compare generated HTML against ground-truth HTML. The graph node embeddings use CLIP features of those components, and the GCN message-passing is trained, but no fitted parameter is renamed as a prediction and no metric is computed from the training fit itself. The evaluation on a self-collected WebSight subset is an in-distribution benchmark choice, not a circular reduction, and the paper also reports results on the external Design2Code benchmark, so the central claim is not self-confirming. The strongest concern raised in review is that the OURS-graph vs OURS-no-graph ablation simultaneously changes graph structure and per-component CLIP feature injection, which is a genuine experimental confound, but a confound in attribution is not a definitional circularity of the kind required for a positive circularity finding. There are no load-bearing self-citations or imported uniqueness theorems, and the limitations section is candid about resource constraints and non-dynamic webpages without asserting a circular derivation. Accordingly, the appropriate finding is no significant circularity, score 0.

Assumptions & free parameters 3 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the component extraction pipeline being accurate, the synthetic WebSight distribution being representative, and several hand-chosen graph construction rules. The paper reports no ablation of these choices and does not provide error bars.

free parameters (3)
  • IoU threshold for edge construction = 0.80
    Section 3.1 sets visual-visual and text-visual edges when IoU exceeds 80%. This hand-chosen threshold is not ablated and directly determines graph connectivity.
  • WebSight training subset size = 20000
    Section 4.1 states the model is trained on a 'curated subset' of 20,000 WebSight pairs. The curation criteria are unspecified, and all central results depend on this subset.
  • Graph encoder and cross-attention hyperparameters
    The number of GCN layers, hidden dimensions, number of visual tokens, learning rate, and model size are not reported anywhere in the preprint, so the fitted configuration cannot be reconstructed.
assumptions (4)
  • standard math The GCN message-passing update defined in Section 2.3 is a correct and suitable encoder for webpage component graphs.
    The paper relies on the standard GCN formulation as background and provides no architectural justification for why this particular graph encoder is sufficient.
  • domain assumption OCR text masking followed by SAM segmentation reliably extracts the salient components of a webpage screenshot.
    Section 3's entire graph construction assumes that the two-step extraction pipeline yields accurate nodes and bounding boxes; no extraction error rates are measured.
  • domain assumption The synthetic WebSight distribution is representative enough to validate Design2Code methods.
    Section 4.1 explicitly avoids real-world data due to instability and limited compute, so the claimed improvements are only demonstrated on synthetic pages.
  • ad hoc to paper Fully connecting all text nodes and using an IoU threshold of 0.8 captures the semantic and spatial relationships needed for HTML generation.
    These edge rules are introduced in Section 3.1 without ablation or comparison to alternative graph constructions, so they are ad hoc choices that the central claim depends on.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Multimodal graph representation learning for website generation based on visual sketch." pith.science (2026). https://pith.science/paper/DZVPSMPG

@misc{pith2026250418729,
  author       = {Pith},
  title        = {Pith review of: Multimodal graph representation learning for website generation based on visual sketch},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/DZVPSMPG}},
  note         = {Machine review of arXiv:2504.18729}
}
read the original abstract

The Design2Code problem, which involves converting digital designs into functional source code, is a significant challenge in software development due to its complexity and time-consuming nature. Traditional approaches often struggle with accurately interpreting the intricate visual details and structural relationships inherent in webpage designs, leading to limitations in automation and efficiency. In this paper, we propose a novel method that leverages multimodal graph representation learning to address these challenges. By integrating both visual and structural information from design sketches, our approach enhances the accuracy and efficiency of code generation, particularly in producing semantically correct and structurally sound HTML code. We present a comprehensive evaluation of our method, demonstrating significant improvements in both accuracy and efficiency compared to existing techniques. Extensive evaluation demonstrates significant improvements of multimodal graph learning over existing techniques, highlighting the potential of our method to revolutionize design-to-code automation. Code available at https://github.com/HySonLab/Design2Code

Figures

Figures reproduced from arXiv: 2504.18729 by the authors.

Figure 1
Figure 1. Overview of the Graph-Enhanced Multimodal Architecture for Generating HTML Code [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. Illustration of the component-extraction pipeline. [PITH_FULL_IMAGE:figures/full_fig_p005_2.png] view at source ↗
Figure 3
Figure 3. Gated Cross Attention Block. The Gated Cross Attention Block integrates three input [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

40 extracted references · 12 canonical work pages

  1. [1]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION format.date year duplicate empty "emp...

  2. [2]

    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 Mensch, Katherine Millican, Malcolm Reynolds, Roman Ring, Eliza Rutherford, Serkan Cabi, Tengda Han, Zhitao Gong, Sina Samangooei, Marianne Monteiro, Jacob Menick, Sebastian Borgeaud, Andrew Brock, Aida Nematzadeh, Sahand Sharifzadeh, Mikolaj Binkow...

    work page 2022

  3. [3]

    Openflamingo: An open-source framework for training large autoregressive vision-language models, 2023

    Anas Awadalla, Irena Gao, Josh Gardner, Jack Hessel, Yusuf Hanafy, Wanrong Zhu, Kalyani Marathe, Yonatan Bitton, Samir Gadre, Shiori Sagawa, Jenia Jitsev, Simon Kornblith, Pang Wei Koh, Gabriel Ilharco, Mitchell Wortsman, and Ludwig Schmidt. Openflamingo: An open-source framework for training large autoregressive vision-language models, 2023. URL https://...

    arXiv 2023

  4. [4]

    pix2code: Generating code from a graphical user interface screenshot

    Tony Beltramelli. pix2code: Generating code from a graphical user interface screenshot. In Proceedings of the ACM SIGCHI Symposium on Engineering Interactive Computing Systems, EICS '18, New York, NY, USA, 2018. Association for Computing Machinery. ISBN 9781450358972. doi:10.1145/3220134.3220135. URL https://doi.org/10.1145/3220134.3220135

    arXiv 2018

  5. [5]

    A novel code generator for graphical user interfaces

    Cai Bo, Jian Luo, and Zhen Feng. A novel code generator for graphical user interfaces. Scientific Reports, 13, 11 2023. doi:10.1038/s41598-023-46500-6

    work page doi:10.1038/s41598-023-46500-6 2023

  6. [6]

    Uniter: Universal image-text representation learning

    Yen-Chun Chen, Linjie Li, Licheng Yu, Ahmed El Kholy, Faisal Ahmed, Zhe Gan, Yu Cheng, and Jingjing Liu. Uniter: Universal image-text representation learning. In Computer Vision – ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXX, pp.\ 104–120, Berlin, Heidelberg, 2020. Springer-Verlag. ISBN 978-3-030-58576-1. doi...

    doi:10.1007/978-3-030-58577-8_7 2020

  7. [7]

    Bert: Pre-training of deep bidirectional transformers for language understanding, 2019

    Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre-training of deep bidirectional transformers for language understanding, 2019. URL https://arxiv.org/abs/1810.04805

    arXiv 2019

  8. [8]

    Vision2ui: A real-world dataset with layout for code generation from ui designs

    Yi Gui, Zhen Li, Yao Wan, Yemin Shi, Hongyu Zhang, Yi Su, Shaoling Dong, Xing Zhou, and Wenbin Jiang. Vision2ui: A real-world dataset with layout for code generation from ui designs. ArXiv, abs/2404.06369, 2024. URL https://api.semanticscholar.org/CorpusID:269010048

    arXiv 2024

Show all 40 references
  1. [9]

    UIC opilot: Automating UI synthesis via hierarchical code generation from webpage designs

    Yi Gui, Yao Wan, Zhen Li, Zhongyi Zhang, Dongping Chen, Hongyu Zhang, Yi Su, Bohua Chen, Xing Zhou, Wenbin Jiang, and Xiangliang Zhang. UIC opilot: Automating UI synthesis via hierarchical code generation from webpage designs. In THE WEB CONFERENCE 2025, 2025. URL https://open...

    2025

  2. [10]

    Images in language space: Exploring the suitability of large language models for vision & language tasks

    Sherzod Hakimov and David Schlangen. Images in language space: Exploring the suitability of large language models for vision & language tasks. In Anna Rogers, Jordan Boyd-Graber, and Naoaki Okazaki (eds.), Findings of the Association for Computational Linguistics: ACL 2023, pp...

    2023 doi

  3. [11]

    Retrieval-enhanced contrastive vision-text models, 2024

    Ahmet Iscen, Mathilde Caron, Alireza Fathi, and Cordelia Schmid. Retrieval-enhanced contrastive vision-text models, 2024. URL https://arxiv.org/abs/2306.07196

    2024 arXiv

  4. [12]

    Perceiver: General perception with iterative attention

    Andrew Jaegle, Felix Gimeno, Andy Brock, Oriol Vinyals, Andrew Zisserman, and Joao Carreira. Perceiver: General perception with iterative attention. In Marina Meila and Tong Zhang (eds.), Proceedings of the 38th International Conference on Machine Learning, volume 139 of Proce...

    2021

  5. [13]

    Graph4gui: Graph neural networks for representing graphical user interfaces

    Yue Jiang, Changkong Zhou, Vikas Garg, and Antti Oulasvirta. Graph4gui: Graph neural networks for representing graphical user interfaces. In Proceedings of the 2024 CHI Conference on Human Factors in Computing Systems, CHI '24, New York, NY, USA, 2024. Association for Computin...

    2024

  6. [14]

    Kipf and Max Welling

    Thomas N. Kipf and Max Welling. Semi-supervised classification with graph convolutional networks. In International Conference on Learning Representations, 2017. URL https://openreview.net/forum?id=SJU4ayYgl

    2017

  7. [15]

    Obelics: An open web-scale filtered dataset of interleaved image-text documents

    Hugo Lauren c on, Lucile Saulnier, Leo Tronchon, Stas Bekman, Amanpreet Singh, Anton Lozhkov, Thomas Wang, Siddharth Karamcheti, Alexander Rush, Douwe Kiela, Matthieu Cord, and Victor Sanh. Obelics: An open web-scale filtered dataset of interleaved image-text documents. In A. ...

    2023

  8. [16]

    Unlocking the conversion of web screenshots into html code with the websight dataset, 2024

    Hugo Laurençon, Léo Tronchon, and Victor Sanh. Unlocking the conversion of web screenshots into html code with the websight dataset, 2024. URL https://arxiv.org/abs/2403.09029

    2024 arXiv

  9. [17]

    Align before fuse: Vision and language representation learning with momentum distillation

    Junnan Li, Ramprasaath Selvaraju, Akhilesh Gotmare, Shafiq Joty, Caiming Xiong, and Steven Chu Hong Hoi. Align before fuse: Vision and language representation learning with momentum distillation. In M. Ranzato, A. Beygelzimer, Y. Dauphin, P.S. Liang, and J. Wortman Vaughan (ed...

    2021

  10. [18]

    BLIP : Bootstrapping language-image pre-training for unified vision-language understanding and generation

    Junnan Li, Dongxu Li, Caiming Xiong, and Steven Hoi. BLIP : Bootstrapping language-image pre-training for unified vision-language understanding and generation. In Kamalika Chaudhuri, Stefanie Jegelka, Le Song, Csaba Szepesvari, Gang Niu, and Sivan Sabato (eds.), Proceedings of...

    2022

  11. [19]

    Visualbert: A simple and performant baseline for vision and language, 2019

    Liunian Harold Li, Mark Yatskar, Da Yin, Cho-Jui Hsieh, and Kai-Wei Chang. Visualbert: A simple and performant baseline for vision and language, 2019. URL https://arxiv.org/abs/1908.03557

    2019 arXiv

  12. [20]

    Vila: On pre-training for visual language models

    Ji Lin, Hongxu Yin, Wei Ping, Pavlo Molchanov, Mohammad Shoeybi, and Song Han. Vila: On pre-training for visual language models. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp.\ 26689--26699, June 2024

    2024

  13. [21]

    Hyperbolic graph neural networks

    Qi Liu, Maximilian Nickel, and Douwe Kiela. Hyperbolic graph neural networks. In H. Wallach, H. Larochelle, A. Beygelzimer, F. d Alch\' e -Buc, E. Fox, and R. Garnett (eds.), Advances in Neural Information Processing Systems, volume 32. Curran Associates, Inc., 2019. URL https...

    2019

  14. [22]

    Lyu, Baishakhi Ray, Abhik Roychoudhury, Shin Hwei Tan, and Patanamon Thongtanunam

    Michael R. Lyu, Baishakhi Ray, Abhik Roychoudhury, Shin Hwei Tan, and Patanamon Thongtanunam. Automatic programming: Large language models and beyond, 2024. URL https://arxiv.org/abs/2405.02213

    2024 arXiv

  15. [23]

    Reverse engineering mobile application user interfaces with remaui (t)

    Tuan Anh Nguyen and Christoph Csallner. Reverse engineering mobile application user interfaces with remaui (t). 2015 30th IEEE/ACM International Conference on Automated Software Engineering (ASE), pp.\ 248--259, 2015. URL https://api.semanticscholar.org/CorpusID:7499368

    2015

  16. [24]

    Learning transferable visual models from natural language supervision

    Alec Radford, Jong Wook Kim, Chris Hallacy, Aditya Ramesh, Gabriel Goh, Sandhini Agarwal, Girish Sastry, Amanda Askell, Pamela Mishkin, Jack Clark, Gretchen Krueger, and Ilya Sutskever. Learning transferable visual models from natural language supervision. In Marina Meila and ...

    2021

  17. [25]

    Recipe for a general, powerful, scalable graph transformer

    Ladislav Ramp\' a s ek, Michael Galkin, Vijay Prakash Dwivedi, Anh Tuan Luu, Guy Wolf, and Dominique Beaini. Recipe for a general, powerful, scalable graph transformer. In S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh (eds.), Advances in Neural Information ...

    2022

  18. [26]

    Sketch2code: Generating a website from a paper mockup, 2019

    Alex Robinson. Sketch2code: Generating a website from a paper mockup, 2019. URL https://arxiv.org/abs/1905.13750

    2019 arXiv

  19. [27]

    Gated graph recurrent neural networks

    Luana Ruiz, Fernando Gama, and Alejandro Ribeiro. Gated graph recurrent neural networks. IEEE Transactions on Signal Processing, 68: 0 6303--6318, 2020. doi:10.1109/TSP.2020.3033962

    2020

  20. [28]

    Sketch2Code , 12 2022

    ShantamVijayputra. Sketch2Code , 12 2022. URL https://www.kaggle.com/datasets/vshantam/sketch2code

    2022

  21. [29]

    Design2code: How far are we from automating front-end engineering?, 2024

    Chenglei Si, Yanzhe Zhang, Zhengyuan Yang, Ruibo Liu, and Diyi Yang. Design2code: How far are we from automating front-end engineering?, 2024. URL https://arxiv.org/abs/2403.03163

    2024 arXiv

  22. [30]

    Learning ui-to-code reverse generator using visual critic without rendering, 2023

    Davit Soselia, Khalid Saifullah, and Tianyi Zhou. Learning ui-to-code reverse generator using visual critic without rendering, 2023. URL https://arxiv.org/abs/2305.14637

    2023 arXiv

  23. [31]

    A survey of neural code intelligence: Paradigms, advances and beyond, 2024

    Qiushi Sun, Zhirui Chen, Fangzhi Xu, Kanzhi Cheng, Chang Ma, Zhangyue Yin, Jianing Wang, Chengcheng Han, Renyu Zhu, Shuai Yuan, Qipeng Guo, Xipeng Qiu, Pengcheng Yin, Xiaoli Li, Fei Yuan, Lingpeng Kong, Xiang Li, and Zhiyong Wu. A survey of neural code intelligence: Paradigms,...

    2024 arXiv

  24. [32]

    LXMERT : Learning cross-modality encoder representations from transformers

    Hao Tan and Mohit Bansal. LXMERT : Learning cross-modality encoder representations from transformers. In Kentaro Inui, Jing Jiang, Vincent Ng, and Xiaojun Wan (eds.), Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th Internation...

    2019 doi

  25. [33]

    Cogvlm: Visual expert for pretrained language models, 2024

    Weihan Wang, Qingsong Lv, Wenmeng Yu, Wenyi Hong, Ji Qi, Yan Wang, Junhui Ji, Zhuoyi Yang, Lei Zhao, Xixuan Song, Jiazheng Xu, Bin Xu, Juanzi Li, Yuxiao Dong, Ming Ding, and Jie Tang. Cogvlm: Visual expert for pretrained language models, 2024. URL https://arxiv.org/abs/2311.03079

    2024 arXiv

  26. [34]

    Webui: A dataset for enhancing visual ui understanding with web semantics

    Jason Wu, Siyan Wang, Siman Shen, Yi-Hao Peng, Jeffrey Nichols, and Jeffrey P Bigham. Webui: A dataset for enhancing visual ui understanding with web semantics. In Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems, CHI '23, New York, NY, USA, 2023. A...

    2023

  27. [35]

    Xing, Xiaodan Liang, and Zhiqiang Shen

    Sukmin Yun, Haokun Lin, Rusiru Thushara, Mohammad Qazim Bhat, Yongxin Wang, Zutao Jiang, Mingkai Deng, Jinhong Wang, Tianhua Tao, Junbo Li, Haonan Li, Preslav Nakov, Timothy Baldwin, Zhengzhong Liu, Eric P. Xing, Xiaodan Liang, and Zhiqiang Shen. Web2code: A large-scale webpag...

    2024 arXiv

  28. [36]

    Large language models meet nl2code: A survey, 2023

    Daoguang Zan, Bei Chen, Fengji Zhang, Dianjie Lu, Bingchao Wu, Bei Guan, Yongji Wang, and Jian-Guang Lou. Large language models meet nl2code: A survey, 2023. URL https://arxiv.org/abs/2212.09420

    2023 arXiv

  29. [37]

    Bridging design and development with automated declarative ui code generation, 2024

    Ting Zhou, Yanjie Zhao, Xinyi Hou, Xiaoyu Sun, Kai Chen, and Haoyu Wang. Bridging design and development with automated declarative ui code generation, 2024. URL https://arxiv.org/abs/2409.11667

    2024 arXiv

  30. [38]

    @esa (Ref

    \@ifxundefined[1] #1\@undefined \@firstoftwo \@secondoftwo \@ifnum[1] #1 \@firstoftwo \@secondoftwo \@ifx[1] #1 \@firstoftwo \@secondoftwo [2] @ #1 \@temptokena #2 #1 @ \@temptokena \@ifclassloaded agu2001 natbib The agu2001 class already includes natbib coding, so you should ...

  31. [39]

    \@lbibitem[] @bibitem@first@sw\@secondoftwo \@lbibitem[#1]#2 \@extra@b@citeb \@ifundefined br@#2\@extra@b@citeb \@namedef br@#2 \@nameuse br@#2\@extra@b@citeb \@ifundefined b@#2\@extra@b@citeb @num @parse #2 @tmp #1 NAT@b@open@#2 NAT@b@shut@#2 \@ifnum @merge>\@ne @bibitem@firs...

  32. [40]

    @open @close @open @close and [1] URL: #1 \@ifundefined chapter * \@mkboth \@ifxundefined @sectionbib * \@mkboth * \@mkboth\@gobbletwo \@ifclassloaded amsart * \@ifclassloaded amsbook * \@ifxundefined @heading @heading NAT@ctr thebibliography [1] @ \@biblabel @NAT@ctr \@bibset...

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.